code
stringlengths
1
1.49M
vector
listlengths
0
7.38k
snippet
listlengths
0
7.38k
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import mekabot.hrl_robot as hr import hrl_lib.util as ut, hrl_lib.transforms as tr import math, numpy as np import copy import sys sys.path.append('../') import compliant_trajectories as ct import segway_motion_calc as smc def compute_workspace(z, hook_angle): firenze = hr.M3HrlRobot(connect=False) rot_mat = tr.Rz(hook_angle)*tr.Rx(math.radians(0))*tr.Ry(math.radians(-90)) delta_list = [math.radians(d) for d in [0.,0.,0.,0.,10.,10.,10.]] x_list,y_list = [],[] if z < -0.4: xmin = 0.10 xmax = 0.65 else: xmin = 0.15 xmax = 0.65 for x in np.arange(xmin,xmax,.01): for y in np.arange(-0.1,-0.50,-0.01): if x<0.3 and y>-0.2: continue q = firenze.IK('right_arm',np.matrix([x,y,z]).T,rot_mat) if q != None and firenze.within_physical_limits_right(q,delta_list)==True: x_list.append(x) y_list.append(y) return np.matrix([x_list,y_list]) def create_workspace_dict(): dd = {} ha_list = [math.radians(d) for d in [0.,90.,-90.]] for ha in ha_list: d = {} for z in np.arange(-0.1,-0.55,-0.01): print 'z:',z pts2d = compute_workspace(z,hook_angle=ha) d[z] = pts2d dd[ha] = {} dd[ha]['pts'] = d ut.save_pickle(dd,'workspace_dict_'+ut.formatted_time()+'.pkl') def create_workspace_boundary(pkl_name): dd = ut.load_pickle(pkl_name) for ha in dd.keys(): pts_dict = dd[ha]['pts'] bndry_dict = {} for z in pts_dict.keys(): print 'z:',z wrkspc = pts_dict[z] if wrkspc.shape[1] < 100: pts_dict.pop(z) continue bndry = smc.compute_boundary(wrkspc) bndry_dict[z] = bndry dd[ha]['bndry'] = bndry_dict ut.save_pickle(dd, pkl_name) create_workspace_dict() #if len(sys.argv) != 2: # print 'Usage:', sys.argv[0], '<wrkspc dict pkl>' # print 'Exiting ...' # sys.exit() #create_workspace_boundary(sys.argv[1])
[ [ 1, 0, 0.2963, 0.0093, 0, 0.66, 0, 4, 0, 1, 0, 0, 4, 0, 0 ], [ 1, 0, 0.3056, 0.0093, 0, 0.66, 0.0909, 775, 0, 2, 0, 0, 775, 0, 0 ], [ 1, 0, 0.3241, 0.0093, 0, 0.66...
[ "import mekabot.hrl_robot as hr", "import hrl_lib.util as ut, hrl_lib.transforms as tr", "import math, numpy as np", "import copy", "import sys", "sys.path.append('../')", "import compliant_trajectories as ct", "import segway_motion_calc as smc", "def compute_workspace(z, hook_angle):\n firenze =...
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import hrl_lib.util as ut import matplotlib_util.util as mpu import math, numpy as np import sys def plot_workspace(pts,ha,z): if pts.shape[1] == 0: return mpu.figure() good_location = pts.mean(1) mpu.plot_yx(pts[1,:].A1,pts[0,:].A1,label='ha:%.1f'%(math.degrees(ha)), axis='equal',linewidth=0) mpu.plot_yx(good_location[1,:].A1,good_location[0,:].A1, axis='equal',linewidth=0,scatter_size=90,color='k') mpu.savefig('z%.2f_ha%.1f.png'%(z,math.degrees(ha))) argv = sys.argv fname = sys.argv[1] dd = ut.load_pickle(fname) color_list = ['b','y','g'] i = 0 mpu.figure(dpi=100) for ha in dd.keys(): d = dd[ha] l = [] key_list = d['pts'].keys() for k in key_list: pts = d['pts'][k] l.append(pts.shape[1]) #plot_workspace(pts,ha,k) ll = zip(key_list,l) ll.sort() key_list,l = zip(*ll) if ha == 0: label = 'Hook Left' elif abs(ha-math.pi/2) < 0.01: label = 'Hook Down' continue else: label = 'Hook Up' mpu.plot_yx(key_list,l,axis=None,label=label, color=color_list[i], xlabel='\# of points with IK soln', ylabel='Height (m)', scatter_size=8) i += 1 max_idx = np.argmax(l) good_height = key_list[max_idx] print 'good_height:', good_height mpu.plot_yx([good_height],[l[max_idx]],axis=None, color='r', xlabel='\# of points with IK soln', ylabel='Height (m)', scatter_size=8) d['height'] = good_height #ut.save_pickle(dd,fname) #mpu.legend() #mpu.savefig('workspace_npts.png') #mpu.show()
[ [ 1, 0, 0.3368, 0.0105, 0, 0.66, 0, 775, 0, 1, 0, 0, 775, 0, 0 ], [ 1, 0, 0.3474, 0.0105, 0, 0.66, 0.0909, 781, 0, 1, 0, 0, 781, 0, 0 ], [ 1, 0, 0.3579, 0.0105, 0, ...
[ "import hrl_lib.util as ut", "import matplotlib_util.util as mpu", "import math, numpy as np", "import sys", "def plot_workspace(pts,ha,z):\n if pts.shape[1] == 0:\n return\n mpu.figure()\n good_location = pts.mean(1)\n mpu.plot_yx(pts[1,:].A1,pts[0,:].A1,label='ha:%.1f'%(math.degrees(ha)...
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import numpy as np import hrl_lib.util as ut #---- X axis ----- #p = ut.load_pickle('eq_pos_2010May01_213254.pkl') #d = ut.load_pickle('stiffness_2010May01_213323.pkl') #n = 0 #--- Y axis ---- #p = ut.load_pickle('eq_pos_2010May01_213832.pkl') #d = ut.load_pickle('stiffness_2010May01_213907.pkl') #n = 1 #--- Z axis p = ut.load_pickle('eq_pos_2010May01_214434.pkl') d = ut.load_pickle('stiffness_2010May01_214512.pkl') n = 2 pos_list = d['pos_list'] force_list = d['force_list'] stiff_list = [] for pos,force in zip(pos_list,force_list): dx = pos[n,0]-p[n,0] f = force[n,0] print 'force:',f print 'dx:',dx stiff_list.append(f/dx) print stiff_list
[ [ 1, 0, 0.4844, 0.0156, 0, 0.66, 0, 954, 0, 1, 0, 0, 954, 0, 0 ], [ 1, 0, 0.5, 0.0156, 0, 0.66, 0.1111, 775, 0, 1, 0, 0, 775, 0, 0 ], [ 14, 0, 0.7188, 0.0156, 0, 0....
[ "import numpy as np", "import hrl_lib.util as ut", "p = ut.load_pickle('eq_pos_2010May01_214434.pkl')", "d = ut.load_pickle('stiffness_2010May01_214512.pkl')", "n = 2", "pos_list = d['pos_list']", "force_list = d['force_list']", "stiff_list = []", "for pos,force in zip(pos_list,force_list):\n dx ...
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import m3.toolbox as m3t import mekabot.hrl_robot as hr import time import math, numpy as np import hrl_lib.util as ut, hrl_lib.transforms as tr def record_initial(firenze): equilibrium_pos_list = [] for i in range(50): equilibrium_pos_list.append(firenze.end_effector_pos('right_arm')) eq_pos = np.column_stack(equilibrium_pos_list).mean(1) ut.save_pickle(eq_pos,'eq_pos_'+ut.formatted_time()+'.pkl') firenze.bias_wrist_ft('right_arm') def record_joint_displacements(): print 'hit ENTER to start the recording process' k=m3t.get_keystroke() pos_list = [] force_list = [] while k == '\r': print 'hit ENTER to record configuration, something else to exit' k=m3t.get_keystroke() firenze.proxy.step() pos_list.append(firenze.end_effector_pos('right_arm')) force_list.append(firenze.get_wrist_force('right_arm', base_frame=True)) ut.save_pickle({'pos_list':pos_list,'force_list':force_list},'stiffness_'+ut.formatted_time()+'.pkl') firenze.stop() if __name__ == '__main__': settings_r = hr.MekaArmSettings(stiffness_list=[0.1939,0.6713,0.997,0.7272,0.75]) firenze = hr.M3HrlRobot(connect = True, right_arm_settings = settings_r) print 'hit a key to power up the arms.' k = m3t.get_keystroke() firenze.power_on() print 'hit a key to test IK' k = m3t.get_keystroke() rot = tr.Ry(math.radians(-90)) p = np.matrix([0.3,-0.40,-0.2]).T firenze.motors_on() #firenze.go_cartesian('right_arm', p, rot) # jep from springloaded door, trial 15 jep = [-0.30365041761032346, 0.3490658503988659, 0.59866827092412689, 1.7924513637028943, 0.4580617747379146, -0.13602429148726047, -0.48610218950666179] firenze.go_jointangles('right_arm', jep) print 'hit a key to record equilibrium position' k = m3t.get_keystroke() record_initial(firenze) record_joint_displacements() print 'hit a key to end everything' k = m3t.get_keystroke() firenze.stop()
[ [ 1, 0, 0.31, 0.01, 0, 0.66, 0, 478, 0, 1, 0, 0, 478, 0, 0 ], [ 1, 0, 0.32, 0.01, 0, 0.66, 0.1429, 4, 0, 1, 0, 0, 4, 0, 0 ], [ 1, 0, 0.33, 0.01, 0, 0.66, 0.2857...
[ "import m3.toolbox as m3t", "import mekabot.hrl_robot as hr", "import time", "import math, numpy as np", "import hrl_lib.util as ut, hrl_lib.transforms as tr", "def record_initial(firenze):\n equilibrium_pos_list = []\n for i in range(50):\n equilibrium_pos_list.append(firenze.end_effector_po...
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import scipy.optimize as so import math, numpy as np import pylab as pl import sys, optparse, time import copy from enthought.mayavi import mlab import mekabot.hrl_robot as hr import mekabot.coord_frames as mcf import matplotlib_util.util as mpu #import util as ut import roslib; roslib.load_manifest('2010_icra_epc_pull') import hrl_lib.util as ut, hrl_lib.transforms as tr import hrl_tilting_hokuyo.display_3d_mayavi as d3m import segway_motion_calc as smc class JointTrajectory(): ''' class to store joint trajectories. data only - use for pickling. ''' def __init__(self): self.time_list = [] # time in seconds self.q_list = [] #each element is a list of 7 joint angles. self.qdot_list = [] #each element is a list of 7 joint angles. self.qdotdot_list = [] #each element is a list of 7 joint angles. ## class to store trajectory of a coord frame executing planar motion (x,y,a) #data only - use for pickling class PlanarTajectory(): def __init__(self): self.time_list = [] # time in seconds self.x_list = [] self.y_list = [] self.a_list = [] class CartesianTajectory(): ''' class to store trajectory of cartesian points. data only - use for pickling ''' def __init__(self): self.time_list = [] # time in seconds self.p_list = [] #each element is a list of 3 coordinates self.v_list = [] #each element is a list of 3 coordinates (velocity) class ForceTrajectory(): ''' class to store time evolution of the force at the end effector. data only - use for pickling ''' def __init__(self): self.time_list = [] # time in seconds self.f_list = [] #each element is a list of 3 coordinates ## # @param traj - JointTrajectory # @return CartesianTajectory after performing FK on traj to compute # cartesian position, velocity def joint_to_cartesian(traj): firenze = hr.M3HrlRobot(connect=False) pts = [] cart_vel = [] for i in range(len(traj.q_list)): q = traj.q_list[i] p = firenze.FK('right_arm', q) pts.append(p.A1.tolist()) if traj.qdot_list != []: qdot = traj.qdot_list[i] jac = firenze.Jac('right_arm', q) vel = jac * np.matrix(qdot).T cart_vel.append(vel.A1[0:3].tolist()) ct = CartesianTajectory() ct.time_list = copy.copy(traj.time_list) ct.p_list = copy.copy(pts) ct.v_list = copy.copy(cart_vel) #return np.matrix(pts).T return ct def plot_forces_quiver(pos_traj,force_traj,color='k'): import arm_trajectories as at #if traj.__class__ == at.JointTrajectory: if isinstance(pos_traj,at.JointTrajectory): pos_traj = joint_to_cartesian(pos_traj) pts = np.matrix(pos_traj.p_list).T label_list = ['X coord (m)', 'Y coord (m)', 'Z coord (m)'] x = pts[0,:].A1.tolist() y = pts[1,:].A1.tolist() forces = np.matrix(force_traj.f_list).T u = (-1*forces[0,:]).A1.tolist() v = (-1*forces[1,:]).A1.tolist() pl.quiver(x,y,u,v,width=0.002,color=color,scale=100.0) # pl.quiver(x,y,u,v,width=0.002,color=color) pl.axis('equal') ## # @param xaxis - x axis for the graph (0,1 or 2) # @param zaxis - for a 3d plot. not implemented. def plot_cartesian(traj, xaxis=None, yaxis=None, zaxis=None, color='b',label='_nolegend_', linewidth=2, scatter_size=10, plot_velocity=False): import arm_trajectories as at #if traj.__class__ == at.JointTrajectory: if isinstance(traj,at.JointTrajectory): traj = joint_to_cartesian(traj) pts = np.matrix(traj.p_list).T label_list = ['X coord (m)', 'Y coord (m)', 'Z coord (m)'] x = pts[xaxis,:].A1.tolist() y = pts[yaxis,:].A1.tolist() if plot_velocity: vels = np.matrix(traj.v_list).T xvel = vels[xaxis,:].A1.tolist() yvel = vels[yaxis,:].A1.tolist() if zaxis == None: mpu.plot_yx(y, x, color, linewidth, '-', scatter_size, label, axis = 'equal', xlabel = label_list[xaxis], ylabel = label_list[yaxis],) if plot_velocity: mpu.plot_quiver_yxv(y, x, np.matrix([xvel,yvel]), width = 0.001, scale = 1.) mpu.legend() else: from numpy import array from enthought.mayavi.api import Engine engine = Engine() engine.start() if len(engine.scenes) == 0: engine.new_scene() z = pts[zaxis,:].A1.tolist() time_list = [t-traj.time_list[0] for t in traj.time_list] mlab.plot3d(x,y,z,time_list,tube_radius=None,line_width=4) mlab.axes() mlab.xlabel(label_list[xaxis]) mlab.ylabel(label_list[yaxis]) mlab.zlabel(label_list[zaxis]) mlab.colorbar(title='Time') # ------------------------------------------- axes = engine.scenes[0].children[0].children[0].children[1] axes.axes.position = array([ 0., 0.]) axes.axes.label_format = '%-#6.2g' axes.title_text_property.font_size=4 ## compute the force that the arm would apply given the stiffness matrix # @param q_actual_traj - Joint Trajectory (actual angles.) # @param q_eq_traj - Joint Trajectory (equilibrium point angles.) # @param torque_traj - JointTrajectory (torques measured at the joints.) # @param rel_stiffness_list - list of 5 elements (stiffness numbers for the joints.) # @return lots of things, look at the code. def compute_forces(q_actual_traj,q_eq_traj,torque_traj,rel_stiffness_list): firenze = hr.M3HrlRobot(connect=False) d_gains_list_mN_deg_sec = [-100,-120,-15,-25,-1.25] d_gains_list = [180./1000.*s/math.pi for s in d_gains_list_mN_deg_sec] stiff_list_mNm_deg = [1800,1300,350,600,60] stiff_list_Nm_rad = [180./1000.*s/math.pi for s in stiff_list_mNm_deg] # stiffness_settings = [0.15,0.7,0.8,0.8,0.8] # dia = np.array(stiffness_settings) * np.array(stiff_list_Nm_rad) dia = np.array(rel_stiffness_list) * np.array(stiff_list_Nm_rad) k_q = np.matrix(np.diag(dia)) dia_inv = 1./dia k_q_inv = np.matrix(np.diag(dia_inv)) actual_cart = joint_to_cartesian(q_actual_traj) eq_cart = joint_to_cartesian(q_eq_traj) force_traj_jacinv = ForceTrajectory() force_traj_stiff = ForceTrajectory() force_traj_torque = ForceTrajectory() k_cart_list = [] for q_actual,q_dot,q_eq,actual_pos,eq_pos,t,tau_m in zip(q_actual_traj.q_list,q_actual_traj.qdot_list,q_eq_traj.q_list,actual_cart.p_list,eq_cart.p_list,q_actual_traj.time_list,torque_traj.q_list): q_eq = firenze.clamp_to_physical_joint_limits('right_arm',q_eq) q_delta = np.matrix(q_actual).T - np.matrix(q_eq).T tau = k_q * q_delta[0:5,0] - np.matrix(np.array(d_gains_list)*np.array(q_dot)[0:5]).T x_delta = np.matrix(actual_pos).T - np.matrix(eq_pos).T jac_full = firenze.Jac('right_arm',q_actual) jac = jac_full[0:3,0:5] jac_full_eq = firenze.Jac('right_arm',q_eq) jac_eq = jac_full_eq[0:3,0:5] k_cart = np.linalg.inv((jac_eq*k_q_inv*jac_eq.T)) # calculating stiff matrix using Jacobian for eq pt. k_cart_list.append(k_cart) pseudo_inv_jac = np.linalg.inv(jac_full*jac_full.T)*jac_full tau_full = np.row_stack((tau,np.matrix(tau_m[5:7]).T)) #force = (-1*pseudo_inv_jac*tau_full)[0:3] force = -1*pseudo_inv_jac[0:3,0:5]*tau force_traj_jacinv.f_list.append(force.A1.tolist()) force_traj_stiff.f_list.append((k_cart*x_delta).A1.tolist()) force_traj_torque.f_list.append((pseudo_inv_jac*np.matrix(tau_m).T)[0:3].A1.tolist()) return force_traj_jacinv,force_traj_stiff,force_traj_torque,k_cart_list ## return two lists containing the radial and tangential components of the forces. # @param f_list - list of forces. (each force is a list of 2 or 3 floats) # @param p_list - list of positions. (each position is a list of 2 or 3 floats) # @param cx - x coord of the center of the circle. # @param cy - y coord of the center of the circle. # @return list of magnitude of radial component, list of magnitude tangential component. def compute_radial_tangential_forces(f_list,p_list,cx,cy): f_rad_l,f_tan_l = [],[] for f,p in zip(f_list,p_list): rad_vec = np.array([p[0]-cx,p[1]-cy]) rad_vec = rad_vec/np.linalg.norm(rad_vec) f_vec = np.array([f[0],f[1]]) f_rad_mag = np.dot(f_vec,rad_vec) f_tan_mag = np.linalg.norm(f_vec-rad_vec*f_rad_mag) f_rad_mag = abs(f_rad_mag) f_rad_l.append(f_rad_mag) f_tan_l.append(f_tan_mag) return f_rad_l,f_tan_l ## find the x and y coord of the center of the circle and the radius that # best matches the data. # @param rad_guess - guess for the radius of the circle # @param x_guess - guess for x coord of center # @param y_guess - guess for y coord of center. # @param pts - 2xN np matrix of points. # @param method - optimization method. ('fmin' or 'fmin_bfgs') # @param verbose - passed onto the scipy optimize functions. whether to print out the convergence info. # @return r,x,y (radius, x and y coord of the center of the circle) def fit_circle(rad_guess,x_guess,y_guess,pts,method,verbose=True): def error_function(params): center = np.matrix((params[0],params[1])).T rad = params[2] #print 'pts.shape', pts.shape #print 'center.shape', center.shape #print 'ut.norm(pts-center).shape',ut.norm(pts-center).shape err = ut.norm(pts-center).A1 - rad res = np.dot(err,err) return res params_1 = [x_guess,y_guess,rad_guess] if method == 'fmin': r = so.fmin(error_function,params_1,xtol=0.0002,ftol=0.000001,full_output=1,disp=verbose) opt_params_1,fopt_1 = r[0],r[1] elif method == 'fmin_bfgs': r = so.fmin_bfgs(error_function, params_1, full_output=1, disp = verbose, gtol=1e-5) opt_params_1,fopt_1 = r[0],r[1] else: raise RuntimeError('unknown method: '+method) params_2 = [x_guess,y_guess+2*rad_guess,rad_guess] if method == 'fmin': r = so.fmin(error_function,params_2,xtol=0.0002,ftol=0.000001,full_output=1,disp=verbose) opt_params_2,fopt_2 = r[0],r[1] elif method == 'fmin_bfgs': r = so.fmin_bfgs(error_function, params_2, full_output=1, disp = verbose, gtol=1e-5) opt_params_2,fopt_2 = r[0],r[1] else: raise RuntimeError('unknown method: '+method) if fopt_2<fopt_1: return opt_params_2[2],opt_params_2[0],opt_params_2[1] else: return opt_params_1[2],opt_params_1[0],opt_params_1[1] ## changes the cartesian trajectory to put everything in the same frame. # NOTE - velocity transformation does not work if the segway is also # moving. This is because I am not logging the velocity of the segway. # @param pts - CartesianTajectory # @param st - object of type PlanarTajectory (segway trajectory) # @return CartesianTajectory def account_segway_motion(cart_traj,st): ct = CartesianTajectory() for i in range(len(cart_traj.p_list)): x,y,a = st.x_list[i], st.y_list[i], st.a_list[i] p_tl = np.matrix(cart_traj.p_list[i]).T p_ts = smc.tsTtl(p_tl, x, y, a) p = p_ts ct.p_list.append(p.A1.tolist()) # this is incorrect. I also need to use the velocity of the # segway. Unclear whether this is useful right now, so not # implementing it. (Advait. Jan 6, 2010.) if cart_traj.v_list != []: v_tl = np.matrix(cart_traj.v_list[i]).T v_ts = smc.tsRtl(v_tl, a) ct.v_list.append(v_ts.A1.tolist()) ct.time_list = copy.copy(cart_traj.time_list) return ct # @param cart_traj - CartesianTajectory # @param z_l - list of zenither heights # @return CartesianTajectory def account_zenithering(cart_traj, z_l): ct = CartesianTajectory() h_start = z_l[0] for i in range(len(cart_traj.p_list)): h = z_l[i] p = cart_traj.p_list[i] p[2] += h - h_start ct.p_list.append(p) # this is incorrect. I also need to use the velocity of the # zenither. Unclear whether this is useful right now, so not # implementing it. (Advait. Jan 6, 2010.) if cart_traj.v_list != []: ct.v_list.append(cart_traj.v_list[i]) ct.time_list = copy.copy(cart_traj.time_list) return ct ## # remove the parts of the trjectory in which the hook is not moving. # @param ct - cartesian trajectory of the end effector in the world frame. # @return 2xN np matrix, reject_idx def filter_cartesian_trajectory(ct): pts_list = ct.p_list ee_start_pos = pts_list[0] l = [pts_list[0]] for i, p in enumerate(pts_list[1:]): l.append(p) pts_2d = (np.matrix(l).T)[0:2,:] st_pt = pts_2d[:,0] end_pt = pts_2d[:,-1] dist_moved = np.linalg.norm(st_pt-end_pt) #if dist_moved < 0.1: if dist_moved < 0.03: reject_idx = i pts_2d = pts_2d[:,reject_idx:] return pts_2d, reject_idx ## # remove the parts of the trjectory in which the hook slipped off # @param ct - cartesian trajectory of the end effector in the world frame. # @param ft - force trajectory # @return cartesian trajectory with the zero force end part removed, force trajectory def filter_trajectory_force(ct, ft): vel_list = copy.copy(ct.v_list) pts_list = copy.copy(ct.p_list) time_list = copy.copy(ct.time_list) ft_list = copy.copy(ft.f_list) f_mag_list = ut.norm(np.matrix(ft.f_list).T).A1.tolist() if len(pts_list) != len(f_mag_list): print 'arm_trajectories.filter_trajectory_force: force and end effector lists are not of the same length.' print 'Exiting ...' sys.exit() n_pts = len(pts_list) i = n_pts - 1 hook_slip_off_threshold = 1.5 # from compliant_trajectories.py while i > 0: if f_mag_list[i] < hook_slip_off_threshold: pts_list.pop() time_list.pop() ft_list.pop() if vel_list != []: vel_list.pop() else: break i -= 1 ct2 = CartesianTajectory() ct2.time_list = time_list ct2.p_list = pts_list ct2.v_list = vel_list ft2 = ForceTrajectory() ft2.time_list = copy.copy(time_list) ft2.f_list = ft_list return ct2, ft2 if __name__ == '__main__': p = optparse.OptionParser() p.add_option('-f', action='store', type='string', dest='fname', help='pkl file to use.', default='') p.add_option('--xy', action='store_true', dest='xy', help='plot the x and y coordinates of the end effector.') p.add_option('--yz', action='store_true', dest='yz', help='plot the y and z coordinates of the end effector.') p.add_option('--xz', action='store_true', dest='xz', help='plot the x and z coordinates of the end effector.') p.add_option('--plot_ellipses', action='store_true', dest='plot_ellipses', help='plot the stiffness ellipse in the x-y plane') p.add_option('--pfc', action='store_true', dest='pfc', help='plot the radial and tangential components of the force.') p.add_option('--pmf', action='store_true', dest='pmf', help='plot things with the mechanism alinged with the axes.') p.add_option('--pff', action='store_true', dest='pff', help='plot the force field corresponding to a stiffness ellipse.') p.add_option('--pev', action='store_true', dest='pev', help='plot the stiffness ellipses for different combinations of the rel stiffnesses.') p.add_option('--plot_forces', action='store_true', dest='plot_forces', help='plot the force in the x-y plane') p.add_option('--plot_forces_error', action='store_true', dest='plot_forces_error', help='plot the error between the computed and measured (ATI) forces in the x-y plane') p.add_option('--xyz', action='store_true', dest='xyz', help='plot in 3d the coordinates of the end effector.') p.add_option('-r', action='store', type='float', dest='rad', help='radius of the joint.', default=None) p.add_option('--noshow', action='store_true', dest='noshow', help='do not display the figure (use while saving figures to disk)') p.add_option('--exptplot', action='store_true', dest='exptplot', help='put all the graphs of an experiment as subplots.') p.add_option('--sturm', action='store_true', dest='sturm', help='make log files to send to sturm') p.add_option('--icra_presentation_plot', action='store_true', dest='icra_presentation_plot', help='plot explaining CEP update.') opt, args = p.parse_args() fname = opt.fname xy_flag = opt.xy yz_flag = opt.yz xz_flag = opt.xz plot_forces_flag = opt.plot_forces plot_ellipses_flag = opt.plot_ellipses plot_forces_error_flag = opt.plot_forces_error plot_force_components_flag = opt.pfc plot_force_field_flag = opt.pff plot_mechanism_frame = opt.pmf xyz_flag = opt.xyz rad = opt.rad show_fig = not(opt.noshow) plot_ellipses_vary_flag = opt.pev expt_plot = opt.exptplot sturm_output = opt.sturm if plot_ellipses_vary_flag: show_fig=False i = 0 ratio_list1 = [0.1,0.3,0.5,0.7,0.9] # coarse search ratio_list2 = [0.1,0.3,0.5,0.7,0.9] # coarse search ratio_list3 = [0.1,0.3,0.5,0.7,0.9] # coarse search # ratio_list1 = [0.7,0.8,0.9,1.0] # ratio_list2 = [0.7,0.8,0.9,1.0] # ratio_list3 = [0.3,0.4,0.5,0.6,0.7] # ratio_list1 = [1.0,2.,3.0] # ratio_list2 = [1.,2.,3.] # ratio_list3 = [0.3,0.4,0.5,0.6,0.7] inv_mean_list,std_list = [],[] x_l,y_l,z_l = [],[],[] s0 = 0.2 #s0 = 0.4 for s1 in ratio_list1: for s2 in ratio_list2: for s3 in ratio_list3: i += 1 s_list = [s0,s1,s2,s3,0.8] #s_list = [s1,s2,s3,s0,0.8] print '################## s_list:', s_list m,s = plot_stiff_ellipse_map(s_list,i) inv_mean_list.append(1./m) std_list.append(s) x_l.append(s1) y_l.append(s2) z_l.append(s3) ut.save_pickle({'x_l':x_l,'y_l':y_l,'z_l':z_l,'inv_mean_list':inv_mean_list,'std_list':std_list}, 'stiff_dict_'+ut.formatted_time()+'.pkl') d3m.plot_points(np.matrix([x_l,y_l,z_l]),scalar_list=inv_mean_list,mode='sphere') mlab.axes() d3m.show() sys.exit() if fname=='': print 'please specify a pkl file (-f option)' print 'Exiting...' sys.exit() d = ut.load_pickle(fname) actual_cartesian_tl = joint_to_cartesian(d['actual']) actual_cartesian = account_segway_motion(actual_cartesian_tl,d['segway']) if d.has_key('zenither_list'): actual_cartesian = account_zenithering(actual_cartesian, d['zenither_list']) eq_cartesian_tl = joint_to_cartesian(d['eq_pt']) eq_cartesian = account_segway_motion(eq_cartesian_tl, d['segway']) if d.has_key('zenither_list'): eq_cartesian = account_zenithering(eq_cartesian, d['zenither_list']) cartesian_force_clean, _ = filter_trajectory_force(actual_cartesian, d['force']) pts_2d, reject_idx = filter_cartesian_trajectory(cartesian_force_clean) if rad != None: #rad = 0.39 # lab cabinet recessed. #rad = 0.42 # kitchen cabinet #rad = 0.80 # lab glass door pts_list = actual_cartesian.p_list eq_pts_list = eq_cartesian.p_list ee_start_pos = pts_list[0] x_guess = ee_start_pos[0] y_guess = ee_start_pos[1] - rad print 'before call to fit_rotary_joint' force_list = d['force'].f_list if sturm_output: str_parts = fname.split('.') sturm_file_name = str_parts[0]+'_sturm.log' print 'Sturm file name:', sturm_file_name sturm_file = open(sturm_file_name,'w') sturm_pts = cartesian_force_clean.p_list print 'len(sturm_pts):', len(sturm_pts) print 'len(pts_list):', len(pts_list) for i,p in enumerate(sturm_pts[1:]): sturm_file.write(" ".join(map(str,p))) sturm_file.write('\n') sturm_file.write('\n') sturm_file.close() rad_guess = rad rad, cx, cy = fit_circle(rad_guess,x_guess,y_guess,pts_2d, method='fmin_bfgs',verbose=False) c_ts = np.matrix([cx, cy, 0.]).T start_angle = tr.angle_within_mod180(math.atan2(pts_2d[0,1]-cy, pts_2d[0,0]-cx) - math.pi/2) end_angle = tr.angle_within_mod180(math.atan2(pts_2d[-1,1]-cy, pts_2d[-1,0]-cx) - math.pi/2) mpu.plot_circle(cx, cy, rad, start_angle, end_angle, label='Actual\_opt', color='r') if opt.icra_presentation_plot: mpu.set_figure_size(30,20) rad = 1.0 x_guess = pts_2d[0,0] y_guess = pts_2d[1,0] - rad rad_guess = rad rad, cx, cy = fit_circle(rad_guess,x_guess,y_guess,pts_2d, method='fmin_bfgs',verbose=False) print 'Estimated rad, cx, cy:', rad, cx, cy start_angle = tr.angle_within_mod180(math.atan2(pts_2d[1,0]-cy, pts_2d[0,0]-cx) - math.pi/2) end_angle = tr.angle_within_mod180(math.atan2(pts_2d[1,-1]-cy, pts_2d[0,-1]-cx) - math.pi/2) subsample_ratio = 1 pts_2d_s = pts_2d[:,::subsample_ratio] cep_force_clean, force_new = filter_trajectory_force(eq_cartesian, d['force']) cep_2d = np.matrix(cep_force_clean.p_list).T[0:2,reject_idx:] # first draw the entire CEP and end effector trajectories mpu.figure() mpu.plot_yx(pts_2d_s[1,:].A1, pts_2d_s[0,:].A1, color='b', label = 'FK', axis = 'equal', alpha = 1.0, scatter_size=7, linewidth=0, marker='x', marker_edge_width = 1.5) cep_2d_s = cep_2d[:,::subsample_ratio] mpu.plot_yx(cep_2d_s[1,:].A1, cep_2d_s[0,:].A1, color='g', label = 'CEP', axis = 'equal', alpha = 1.0, scatter_size=10, linewidth=0, marker='+', marker_edge_width = 1.5) mpu.plot_circle(cx, cy, rad, start_angle, end_angle, label='Estimated Kinematics', color='r', alpha=0.7) mpu.plot_radii(cx, cy, rad, start_angle, end_angle, interval=end_angle-start_angle, color='r', alpha=0.7) mpu.legend() mpu.savefig('one.png') # now zoom in to a small region to show the force # decomposition. zoom_location = 10 pts_2d_zoom = pts_2d[:,:zoom_location] cep_2d_zoom = cep_2d[:,:zoom_location] mpu.figure() mpu.plot_yx(pts_2d_zoom[1,:].A1, pts_2d_zoom[0,:].A1, color='b', label = 'FK', axis = 'equal', alpha = 1.0, scatter_size=7, linewidth=0, marker='x', marker_edge_width = 1.5) mpu.plot_yx(cep_2d_zoom[1,:].A1, cep_2d_zoom[0,:].A1, color='g', label = 'CEP', axis = 'equal', alpha = 1.0, scatter_size=10, linewidth=0, marker='+', marker_edge_width = 1.5) mpu.pl.xlim(0.28, 0.47) mpu.legend() mpu.savefig('two.png') rad, cx, cy = fit_circle(1.0,x_guess,y_guess,pts_2d_zoom, method='fmin_bfgs',verbose=False) print 'Estimated rad, cx, cy:', rad, cx, cy start_angle = tr.angle_within_mod180(math.atan2(pts_2d[1,0]-cy, pts_2d[0,0]-cx) - math.pi/2) end_angle = tr.angle_within_mod180(math.atan2(pts_2d_zoom[1,-1]-cy, pts_2d_zoom[0,-1]-cx) - math.pi/2) mpu.plot_circle(cx, cy, rad, start_angle, end_angle, label='Estimated Kinematics', color='r', alpha=0.7) mpu.pl.xlim(0.28, 0.47) mpu.legend() mpu.savefig('three.png') current_pos = pts_2d_zoom[:,-1] radial_vec = current_pos - np.matrix([cx,cy]).T radial_vec = radial_vec / np.linalg.norm(radial_vec) tangential_vec = np.matrix([[0,-1],[1,0]]) * radial_vec mpu.plot_quiver_yxv([pts_2d_zoom[1,-1]], [pts_2d_zoom[0,-1]], radial_vec, scale=10., width = 0.002) rad_text_loc = pts_2d_zoom[:,-1] + np.matrix([0.001,0.01]).T mpu.pl.text(rad_text_loc[0,0], rad_text_loc[1,0], '\huge{$\hat v_{rad}$}') mpu.plot_quiver_yxv([pts_2d_zoom[1,-1]], [pts_2d_zoom[0,-1]], tangential_vec, scale=10., width = 0.002) tan_text_loc = pts_2d_zoom[:,-1] + np.matrix([-0.012, -0.011]).T mpu.pl.text(tan_text_loc[0,0], tan_text_loc[1,0], s = '\huge{$\hat v_{tan}$}') mpu.pl.xlim(0.28, 0.47) mpu.legend() mpu.savefig('four.png') wrist_force = -np.matrix(force_new.f_list[zoom_location]).T frad = (wrist_force[0:2,:].T * radial_vec)[0,0] * radial_vec mpu.plot_quiver_yxv([pts_2d_zoom[1,-1]], [pts_2d_zoom[0,-1]], wrist_force, scale=50., width = 0.002, color='y') wf_text = rad_text_loc + np.matrix([-0.05,0.015]).T mpu.pl.text(wf_text[0,0], wf_text[1,0], color='y', fontsize = 15, s = 'Wrist Force') mpu.plot_quiver_yxv([pts_2d_zoom[1,-1]], [pts_2d_zoom[0,-1]], frad, scale=50., width = 0.002, color='y') frad_text = rad_text_loc + np.matrix([0.,0.015]).T mpu.pl.text(frad_text[0,0], frad_text[1,0], color='y', s = '\huge{$\hat F_{rad}$}') mpu.pl.xlim(0.28, 0.47) mpu.legend() mpu.savefig('five.png') frad = (wrist_force[0:2,:].T * radial_vec)[0,0] hook_force_motion = -(frad - 5) * radial_vec * 0.001 tangential_motion = 0.01 * tangential_vec total_cep_motion = hook_force_motion + tangential_motion mpu.plot_quiver_yxv([cep_2d_zoom[1,-1]], [cep_2d_zoom[0,-1]], hook_force_motion, scale=0.1, width = 0.002) hw_text = cep_2d_zoom[:,-1] + np.matrix([-0.002,-0.012]).T mpu.pl.text(hw_text[0,0], hw_text[1,0], color='k', fontsize=14, s = '$h[t]$ = $0.1cm/N \cdot (|\hat{F}_{rad}|-5N) \cdot \hat{v}_{rad}$') mpu.pl.xlim(0.28, 0.47) mpu.legend() mpu.savefig('six.png') mpu.plot_quiver_yxv([cep_2d_zoom[1,-1]], [cep_2d_zoom[0,-1]], tangential_motion, scale=0.1, width = 0.002) mw_text = cep_2d_zoom[:,-1] + np.matrix([-0.038,0.001]).T mpu.pl.text(mw_text[0,0], mw_text[1,0], color='k', fontsize=14, s = '$m[t]$ = $1cm \cdot \hat{v}_{tan}$') mpu.pl.xlim(0.28, 0.47) mpu.legend() mpu.savefig('seven.png') mpu.plot_quiver_yxv([cep_2d_zoom[1,-1]], [cep_2d_zoom[0,-1]], total_cep_motion, scale=0.1, width = 0.002) cep_text = cep_2d_zoom[:,-1] + np.matrix([-0.058,-0.013]).T mpu.pl.text(cep_text[0,0], cep_text[1,0], color='k', fontsize=14, s = '$x_{eq}[t]$ = &x_{eq}[t-1] + m[t] + h[t]$') mpu.pl.xlim(0.28, 0.47) mpu.legend() mpu.savefig('eight.png') new_cep = cep_2d_zoom[:,-1] + total_cep_motion mpu.plot_yx(new_cep[1,:].A1, new_cep[0,:].A1, color='g', axis = 'equal', alpha = 1.0, scatter_size=10, linewidth=0, marker='+', marker_edge_width = 1.5) mpu.pl.xlim(0.28, 0.47) mpu.legend() mpu.savefig('nine.png') #mpu.plot_radii(cx, cy, rad, start_angle, end_angle, # interval=end_angle-start_angle, color='r', # alpha=0.7) if plot_mechanism_frame: if expt_plot: pl.subplot(231) # transform points so that the mechanism is in a fixed position. start_pt = actual_cartesian.p_list[0] x_diff = start_pt[0] - cx y_diff = start_pt[1] - cy angle = math.atan2(y_diff,x_diff) - math.radians(90) rot_mat = tr.Rz(angle)[0:2,0:2] translation_mat = np.matrix([cx,cy]).T robot_width,robot_length = 0.1,0.2 robot_x_list = [-robot_width/2,-robot_width/2,robot_width/2,robot_width/2,-robot_width/2] robot_y_list = [-robot_length/2,robot_length/2,robot_length/2,-robot_length/2,-robot_length/2] robot_mat = rot_mat*(np.matrix([robot_x_list,robot_y_list]) - translation_mat) mpu.plot_yx(robot_mat[1,:].A1,robot_mat[0,:].A1,linewidth=2,scatter_size=0, color='k',label='torso', axis='equal') pts2d_actual = (np.matrix(actual_cartesian.p_list).T)[0:2] pts2d_actual_t = rot_mat *(pts2d_actual - translation_mat) mpu.plot_yx(pts2d_actual_t[1,:].A1,pts2d_actual_t[0,:].A1,scatter_size=20,label='FK', axis = 'equal') end_pt = pts2d_actual_t[:,-1] end_angle = tr.angle_within_mod180(math.atan2(end_pt[1,0],end_pt[0,0])-math.radians(90)) mpu.plot_circle(0,0,rad,0.,end_angle,label='Actual_opt',color='r') mpu.plot_radii(0,0,rad,0.,end_angle,interval=math.radians(15),color='r') pl.legend(loc='best') pl.axis('equal') if not(expt_plot): str_parts = fname.split('.') fig_name = str_parts[0]+'_robot_pose.png' pl.savefig(fig_name) pl.figure() else: pl.subplot(232) pl.text(0.1,0.15,d['info']) pl.text(0.1,0.10,'control: '+d['strategy']) pl.text(0.1,0.05,'robot angle: %.2f'%math.degrees(angle)) pl.text(0.1,0,'optimized radius: %.2f'%rad_opt) pl.text(0.1,-0.05,'radius used: %.2f'%rad) pl.text(0.1,-0.10,'opening angle: %.2f'%math.degrees(end_angle)) s_list = d['stiffness'].stiffness_list s_scale = d['stiffness'].stiffness_scale sl = [min(s*s_scale,1.0) for s in s_list] pl.text(0.1,-0.15,'stiffness list: %.2f, %.2f, %.2f, %.2f'%(sl[0],sl[1],sl[2],sl[3])) pl.text(0.1,-0.20,'stop condition: '+d['result']) time_dict = d['time_dict'] pl.text(0.1,-0.25,'time to hook: %.2f'%(time_dict['before_hook']-time_dict['before_pull'])) pl.text(0.1,-0.30,'time to pull: %.2f'%(time_dict['before_pull']-time_dict['after_pull'])) pl.ylim(-0.45,0.25) if not(expt_plot): pl.figure() if xy_flag: st_pt = pts_2d[:,0] end_pt = pts_2d[:,-1] # if rad != None: # start_angle = tr.angle_within_mod180(math.atan2(st_pt[1,0]-cy,st_pt[0,0]-cx) - math.radians(90)) # end_angle = tr.angle_within_mod180(math.atan2(end_pt[1,0]-cy,end_pt[0,0]-cx) - math.radians(90)) # # print 'start_angle, end_angle:', math.degrees(start_angle), math.degrees(end_angle) # print 'angle through which mechanism turned:', math.degrees(end_angle-start_angle) if expt_plot: pl.subplot(233) plot_cartesian(actual_cartesian, xaxis=0, yaxis=1, color='b', label='FK', plot_velocity=False) plot_cartesian(eq_cartesian, xaxis=0,yaxis=1,color='g',label='Eq Point') #leg = pl.legend(loc='best',handletextsep=0.020,handlelen=0.003,labelspacing=0.003) #leg.draw_frame(False) elif yz_flag: plot_cartesian(actual_cartesian,xaxis=1,yaxis=2,color='b',label='FK') plot_cartesian(eq_cartesian, xaxis=1,yaxis=2,color='g',label='Eq Point') elif xz_flag: plot_cartesian(actual_cartesian,xaxis=0,yaxis=2,color='b',label='FK') plot_cartesian(eq_cartesian, xaxis=0,yaxis=2,color='g',label='Eq Point') if plot_forces_flag or plot_forces_error_flag or plot_ellipses_flag or plot_force_components_flag or plot_force_field_flag: arm_stiffness_list = d['stiffness'].stiffness_list scale = d['stiffness'].stiffness_scale asl = [min(scale*s,1.0) for s in arm_stiffness_list] ftraj_jinv,ftraj_stiff,ftraj_torque,k_cart_list=compute_forces(d['actual'],d['eq_pt'], d['torque'],asl) if plot_forces_flag: plot_forces_quiver(actual_cartesian,d['force'],color='k') #plot_forces_quiver(actual_cartesian,ftraj_jinv,color='y') #plot_forces_quiver(actual_cartesian,ftraj_stiff,color='y') if plot_ellipses_flag: #plot_stiff_ellipses(k_cart_list,actual_cartesian) if expt_plot: subplotnum=234 else: pl.figure() subplotnum=111 plot_stiff_ellipses(k_cart_list,eq_cartesian,subplotnum=subplotnum) if plot_forces_error_flag: plot_error_forces(d['force'].f_list,ftraj_jinv.f_list) #plot_error_forces(d['force'].f_list,ftraj_stiff.f_list) if plot_force_components_flag: p_list = actual_cartesian.p_list cx = 45. cy = -0.3 frad_list,ftan_list = compute_radial_tangential_forces(d['force'].f_list,p_list,cx,cy) if expt_plot: pl.subplot(235) else: pl.figure() time_list = d['force'].time_list time_list = [t-time_list[0] for t in time_list] x_coord_list = np.matrix(p_list)[:,0].A1.tolist() mpu.plot_yx(frad_list,x_coord_list,scatter_size=50,color=time_list,cb_label='time',axis=None) pl.xlabel('x coord of end effector (m)') pl.ylabel('magnitude of radial force (N)') pl.title(d['info']) if expt_plot: pl.subplot(236) else: pl.figure() mpu.plot_yx(ftan_list,x_coord_list,scatter_size=50,color=time_list,cb_label='time',axis=None) pl.xlabel('x coord of end effector (m)') pl.ylabel('magnitude of tangential force (N)') pl.title(d['info']) if plot_force_field_flag: plot_stiffness_field(k_cart_list[0],plottitle='start') plot_stiffness_field(k_cart_list[-1],plottitle='end') str_parts = fname.split('.') if d.has_key('strategy'): addon = '' if opt.xy: addon = '_xy' if opt.xz: addon = '_xz' fig_name = str_parts[0]+'_'+d['strategy']+addon+'.png' else: fig_name = str_parts[0]+'_res.png' if expt_plot: f = pl.gcf() curr_size = f.get_size_inches() f.set_size_inches(curr_size[0]*2,curr_size[1]*2) f.savefig(fig_name) if show_fig: pl.show() else: print '################################' print 'show_fig is FALSE' if not(expt_plot): pl.savefig(fig_name) if xyz_flag: plot_cartesian(traj, xaxis=0,yaxis=1,zaxis=2) mlab.show()
[ [ 1, 0, 0.0347, 0.0011, 0, 0.66, 0, 359, 0, 1, 0, 0, 359, 0, 0 ], [ 1, 0, 0.0369, 0.0011, 0, 0.66, 0.0357, 526, 0, 2, 0, 0, 526, 0, 0 ], [ 1, 0, 0.038, 0.0011, 0, 0...
[ "import scipy.optimize as so", "import math, numpy as np", "import pylab as pl", "import sys, optparse, time", "import copy", "from enthought.mayavi import mlab", "import mekabot.hrl_robot as hr", "import mekabot.coord_frames as mcf", "import matplotlib_util.util as mpu", "import roslib; roslib.lo...
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import sys, time, os, optparse import math, numpy as np import copy import mekabot.coord_frames as mcf import compliant_trajectories as ct import m3.toolbox as m3t import roslib; roslib.load_manifest('2010_icra_epc_pull') import hrl_lib.util as ut import hrl_lib.transforms as tr if __name__=='__main__': p = optparse.OptionParser() p.add_option('--ha', action='store', dest='ha',type='float', default=None,help='hook angle (degrees).') p.add_option('--ft', action='store', dest='ft',type='float', default=80.,help='force threshold (Newtons). [default 80.]') p.add_option('--info', action='store', type='string', dest='info_string', help='string to save in the pkl log.', default='') p.add_option('--pull_fixed', action='store_true', dest='pull_fixed', help='pull with the segway stationary') p.add_option('--lead', action='store_true', dest='lead', help='move the segway while pulling') p.add_option('--lpi', action='store_true', dest='lpi', help='use the laser pointer interface to designate hooking location') p.add_option('-p', action='store', dest='p',type='int', default=2,help='position number') p.add_option('-z', action='store', dest='z',type='float', default=1.0,help='zenither height') p.add_option('--sa', action='store', dest='sa',type='float', default=0.0,help='servo angle at which to take camera image (DEGREES)') p.add_option('--sliding_left', action='store_true', dest='sliding_left', help='defining the initial motion of the hook.') p.add_option('--use_jacobian', action='store_true', dest='use_jacobian', help='assume that kinematics estimation gives a jacobian') opt, args = p.parse_args() ha = opt.ha z = opt.z sa = opt.sa ft = opt.ft info_string = opt.info_string lead_flag = opt.lead lpi_flag = opt.lpi pull_fixed_flag = opt.pull_fixed move_segway_flag = not pull_fixed_flag pnum = opt.p arm = 'right_arm' try: if ha == None: print 'please specify hook angle (--ha)' print 'Exiting...' sys.exit() cmg = ct.CompliantMotionGenerator(move_segway = move_segway_flag, use_right_arm = True, use_left_arm = False) if lead_flag: sys.path.append(os.environ['HRLBASEPATH']+'/src/projects/lead') import lead print 'hit a key to start leading.' k=m3t.get_keystroke() cmg.firenze.power_on() if move_segway_flag: mec = cmg.segway_command_node else: import segway_omni.segway as segway mec = segway.Mecanum() zed = cmg.z import mekabot.hrl_robot as hr settings_lead = hr.MekaArmSettings(stiffness_list=[0.2,0.3,0.3,0.5,0.8]) cmg.firenze.set_arm_settings(settings_lead,None) follower = lead.Lead(cmg.firenze,'right_arm',mec,zed, max_allowable_height=zed.calib['max_height'], init_height = zed.get_position_meters()) qr = [0, 0, 0, math.pi/2, -(math.radians(ha)-ct.hook_3dprint_angle), 0, 0] follower.start_lead_thread(qr=qr) print 'hit a key to start hook and pull.' k=m3t.get_keystroke() follower.stop() cmg.firenze.set_arm_settings(cmg.settings_stiff,None) cmg.firenze.step() cmg.firenze.pose_robot('right_arm') print 'Hit ENTER to reposition the robot' k=m3t.get_keystroke() if k!='\r': print 'You did not press ENTER.' print 'Exiting ...' sys.exit() hook_location = cmg.firenze.end_effector_pos('right_arm') pull_loc = cmg.reposition_robot(hook_location) elif lpi_flag: import lpi # import point_cloud_features.pointcloud_features as ppf import hrl_tilting_hokuyo.processing_3d as p3d # pc_feat = ppf.PointcloudFeatureExtractor(ros_init_node=False) cmg.z.torque_move_position(1.0) cmg.firenze.power_on() cmg.movement_posture() if z<0.5: print 'desired zenither height of %.2f is unsafe.'%(z) print 'Exiting...' sys.exit() hook_location = None #z_list = [1.0,0.5] z_list = [opt.z] i = 0 while hook_location == None: if i == len(z_list): print 'Did not get a click. Exiting...' sys.exit() z = z_list[i] cmg.z.torque_move_position(z) hook_location = lpi.select_location(cmg.cam,cmg.thok,math.radians(sa)) i += 1 hl_thok0 = mcf.thok0Tglobal(hook_location) hl_torso = mcf.torsoTglobal(hook_location) t_begin = time.time() angle = 0. pull_loc,residual_angle = cmg.reposition_robot(hl_torso,angle,math.radians(ha), position_number=pnum) print 'pull_loc:',pull_loc.A1.tolist() starting_location = copy.copy(hl_torso) starting_angle = -angle pose_dict = {} pose_dict['loc'] = starting_location pose_dict['angle'] = angle pose_dict['residual_angle'] = residual_angle pose_dict['position_number'] = pnum if opt.sliding_left: thresh = 2. else: thresh = 5. res, jep = cmg.search_and_hook(arm, math.radians(ha), pull_loc, residual_angle, thresh) print 'result of search and hook:', res if res != 'got a hook': print 'Did not get a hook.' print 'Exiting...' sys.exit() elif pull_fixed_flag: print 'hit a key to power up the arms.' k=m3t.get_keystroke() cmg.firenze.power_on() t_begin = time.time() pull_loc = np.matrix([0.55, -0.2, -.23]).T if opt.sliding_left: thresh = 2. else: thresh = 5. res, jep = cmg.search_and_hook(arm, math.radians(ha), pull_loc, 0., thresh) print 'result of search and hook:', res if res != 'got a hook': print 'Did not get a hook.' print 'Exiting...' sys.exit() residual_angle = 0. pose_dict = {} else: raise RuntimeError('Unsupported. Advait Jan 02, 2010.') if opt.use_jacobian: kinematics_estimation = 'jacobian' else: kinematics_estimation = 'rotation_center' t_pull = time.time() cmg.pull(arm, math.radians(ha), residual_angle, ft, jep, strategy = 'control_radial_force', info_string=info_string, cep_vel = 0.10, kinematics_estimation=kinematics_estimation, pull_left = opt.sliding_left) t_end = time.time() pose_dict['t0'] = t_begin pose_dict['t1'] = t_pull pose_dict['t2'] = t_end ut.save_pickle(pose_dict,'pose_dict_'+ut.formatted_time()+'.pkl') print 'hit a key to end everything' k=m3t.get_keystroke() cmg.stop() except: # catch all exceptions, stop and re-raise them print 'Hello, Mr. Exception' cmg.stop() raise
[ [ 1, 0, 0.1301, 0.0041, 0, 0.66, 0, 509, 0, 4, 0, 0, 509, 0, 0 ], [ 1, 0, 0.1341, 0.0041, 0, 0.66, 0.1, 526, 0, 2, 0, 0, 526, 0, 0 ], [ 1, 0, 0.1382, 0.0041, 0, 0.6...
[ "import sys, time, os, optparse", "import math, numpy as np", "import copy", "import mekabot.coord_frames as mcf", "import compliant_trajectories as ct", "import m3.toolbox as m3t", "import roslib; roslib.load_manifest('2010_icra_epc_pull')", "import roslib; roslib.load_manifest('2010_icra_epc_pull')"...
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import roslib roslib.load_manifest('equilibrium_point_control') import numpy as np, math import scipy.optimize as so import scipy.ndimage as ni import matplotlib_util.util as mpu import hrl_lib.util as ut import hrl_lib.transforms as tr import hrl_hokuyo.hokuyo_processing as hp import mekabot.coord_frames as mcf import util as uto from opencv.highgui import * ## # @param pts - 2xN np matrix # @return r,theta (two 1D np arrays) def cartesian_to_polar(pts): r = ut.norm(pts).A1 theta = np.arctan2(pts[1,:],pts[0,:]).A1 return r,theta ## # @param r - 1D np array # @param theta - 1D np array (RADIANS) # @return 2xN np matrix of cartesian points def polar_to_cartesian(r,theta): x = r*np.cos(theta) y = r*np.sin(theta) return np.matrix(np.row_stack((x,y))) ## # mx,my,ma - motion of the robot # cx,cy - axis of mechanism in robot frame. # start_angle,end_angle - between 0 and 2*pi def point_contained(mx,my,ma,cx,cy,rad,pts,start_angle,end_angle,buffer): if abs(mx)>0.2 or abs(my)>0.2 or abs(ma)>math.radians(40): # print 'too large a motion for point_contained' return np.array([[]]) pts_t = pts + np.matrix([mx,my]).T pts_t = tr.Rz(-ma)[0:2,0:2]*pts_t r,t = cartesian_to_polar(pts_t-np.matrix([cx,cy]).T) t = np.mod(t,math.pi*2) # I want theta to be b/w 0 and 2pi if start_angle<end_angle: f = np.row_stack((r<rad+buffer,r>rad-buffer/2.,t<end_angle,t>start_angle)) else: f = np.row_stack((r<rad+buffer,r>rad-buffer/2.,t<start_angle,t>end_angle)) idxs = np.where(np.all(f,0)) r_filt = r[idxs] t_filt = t[idxs] return polar_to_cartesian(r_filt,t_filt)+np.matrix([cx,cy]).T def optimize_position(cx,cy,rad,curr_pos,eq_pos,pts,bndry,start_angle, end_angle,buffer,tangential_force): scale_x,scale_y,scale_a = 1.,1.,1. b = min(abs(tangential_force),60.) if end_angle>start_angle: # min_alpha = math.radians(30) max_alpha = math.radians(90) else: # min_alpha = math.radians(-150) max_alpha = math.radians(-90) min_alpha = max_alpha - math.radians(60-b*0.7) dist_moved_weight = 0.4 - 0.3*b/60. alpha_weight = 0.4+1.0*b/60. bndry_weight = 1. pts_in_weight = 1. print 'OPTIMIZE_POSITION' print 'start_angle:', math.degrees(start_angle) print 'end_angle:', math.degrees(end_angle) print 'tangential_force:', tangential_force def error_function(params): mx,my,ma = params[0],params[1],params[2] mx,my,ma = mx/scale_x,my/scale_y,ma/scale_a #x,y = params[0],params[1] pts_in = point_contained(mx,my,ma,cx,cy,rad,pts, start_angle,end_angle,buffer) p = tr.Rz(ma)*curr_pos-np.matrix([mx,my,0.]).T p_eq = tr.Rz(ma)*eq_pos-np.matrix([mx,my,0.]).T dist_moved = math.sqrt(mx*mx+my*my)+abs(ma)*0.2 dist_bndry = dist_from_boundary(p_eq,bndry,pts) alpha = math.pi-(start_angle-ma) if alpha<min_alpha: alpha_cost = min_alpha-alpha elif alpha>max_alpha: alpha_cost = alpha-max_alpha else: alpha_cost = 0. alpha_cost = alpha_cost * alpha_weight move_cost = dist_moved * dist_moved_weight bndry_cost = dist_bndry * bndry_weight pts_in_cost = pts_in.shape[1]/1000. * pts_in_weight # print '---------------------------------' # print 'alpha:',math.degrees(alpha) # print 'alpha_cost:',alpha_cost # print 'mx,my:',mx,my # print 'dist_moved:',dist_moved # print 'dist_bndry:',dist_bndry # print 'pts_in.shape[1]:',pts_in.shape[1] # print 'move_cost:', move_cost # print 'bndry_cost:',bndry_cost # print 'pts_in_cost:',pts_in_cost # return -pts_in.shape[1]+dist_moved - bndry_cost err = -pts_in_cost-bndry_cost+move_cost+alpha_cost # print 'error function value:',err return err params_1 = [0.,0.,0.] res = so.fmin_bfgs(error_function,params_1,full_output=1) r,f = res[0],res[1] # r,f,d = so.fmin_l_bfgs_b(error_function,params_1,approx_grad=True, # bounds=[(-0.1*scale_x,0.1*scale_x), # (-0.1*scale_y,0.1*scale_y), # (-math.radians(15)*scale_a, # math.radians(15)*scale_a)], # m=10, factr=10000000.0, # pgtol=1.0000000000000001e-05, # epsilon=0.0001, iprint=-1, # maxfun=1000) opt_params = r # print 'optimized value:',f mx,my,ma = opt_params[0]/scale_x,opt_params[1]/scale_y,\ opt_params[2]/scale_a error_function(opt_params) return mx,my,ma #return opt_params[0],opt_params[1] ## # compute the boundary of the 2D points. Making assumptions about # the density of the points, tested with workspace_dict only. # @param pts - 2xN np matrix def compute_boundary(pts): npim1,nx,ny,br = hp.xy_map_to_np_image(pts,m_per_pixel=0.01,dilation_count=0,padding=10) npim1 = npim1/255 npim = np.zeros(npim1.shape,dtype='int') npim[:,:] = npim1[:,:] connect_structure = np.empty((3,3),dtype='int') connect_structure[:,:] = 1 erim = ni.binary_erosion(npim,connect_structure,iterations=1) bim = npim-erim tup = np.where(bim>0) bpts = np.row_stack((nx-tup[0],ny-tup[1]))*0.01 + br # cvim = uto.np2cv(bim) # cvSaveImage('boundary.png',cvim) return np.matrix(bpts) ## #return 2x1 vector from closest boundary point def vec_from_boundary(curr_pos,bndry): p = curr_pos[0:2,:] v = p-bndry min_idx = np.argmin(ut.norm(v)) return v[:,min_idx] ## #return distance from boundary. (-ve if outside the boundary) # @param curr_pos - can be 3x1 np matrix # @param bndry - boundary (2xN np matrix) # @param pts - 2xN np matrix. pts whose boundary is bndry def dist_from_boundary(curr_pos,bndry,pts): mv = vec_from_boundary(curr_pos,bndry) # spoly = sg.Polygon((bndry.T).tolist()) # spt = sg.Point(curr_pos[0,0],curr_pos[1,0]) d = np.linalg.norm(mv) p = curr_pos[0:2,:] v = p-pts min_dist = np.min(ut.norm(v)) # print 'min_dist,d:',min_dist,d # print 'min_dist >= d',min_dist >= d-0.001 if min_dist >= d-0.001: # print 'I predict outside workspace' d = -d # if spoly.contains(spt) == False: # print 'Shapely predicts outside workspace' # d = -d return d ## # @param curr_pos - current location of end effector. 3x1 np matrix # @param bndry - workspace boundary. 2xN np matrix def close_to_boundary(curr_pos,bndry,pts,dist_thresh): min_dist = dist_from_boundary(curr_pos,bndry,pts) return min_dist <= dist_thresh def visualize_boundary(): d = ut.load_pickle('../../pkls/workspace_dict_2009Sep03_010426.pkl') z = -0.23 k = d.keys() k_idx = np.argmin(np.abs(np.array(k)-z)) pts = d[k[k_idx]] bpts = compute_boundary(pts) cl_list = [] for pt in pts.T: if close_to_boundary(pt.T,bpts,dist_thresh=0.05)==True: cl_list.append(pt.A1.tolist()) clpts = np.matrix(cl_list).T print 'clpts.shape:', clpts.shape mpu.plot_yx(pts[1,:].A1,pts[0,:].A1,linewidth=0) mpu.plot_yx(clpts[1,:].A1,clpts[0,:].A1,linewidth=0,color='r') mpu.plot_yx(bpts[1,:].A1,bpts[0,:].A1,linewidth=0,color='y') mpu.show() ## transform from torso start to torso local frame. # @param pts - 3xN np matrix in ts coord frame. # @param x,y,a - motion of the segway (in the ms frame) # @return pts_tl def tlTts(pts_ts,x,y,a): pts_ms = mcf.mecanumTglobal(mcf.globalTtorso(pts_ts)) v_org_ms = np.matrix([x,y,0.]).T pts_ml = tr.Rz(a)*(pts_ms-v_org_ms) pts_tl = mcf.torsoTglobal(mcf.globalTmecanum(pts_ml)) return pts_tl ## transform from torso local to torso start frame. # @param pts - 3xN np matrix in tl coord frame. # @param x,y,a - motion of the segway (in the ms frame) # @return pts_ts def tsTtl(pts_tl,x,y,a): pts_ml = mcf.mecanumTglobal(mcf.globalTtorso(pts_tl)) v_org_ms = np.matrix([x,y,0.]).T pts_ms = tr.Rz(-a) * pts_ml + v_org_ms pts_ts = mcf.torsoTglobal(mcf.globalTmecanum(pts_ms)) return pts_ts ## rotate vector from torso local to torso start frame. # @param vecs_tl - 3xN np matrix in tl coord frame. # @param a - motion of the segway (in the ms frame) # @return vecs_ts def tsRtl(vecs_tl, a): vecs_ml = mcf.mecanumTglobal(mcf.globalTtorso(vecs_tl, True), True) vecs_ms = tr.Rz(-a) * vecs_ml vecs_ts = mcf.torsoTglobal(mcf.globalTmecanum(vecs_ms, True), True) return vecs_ts ## rotate vector from torso local to torso start frame. # @param vecs_tl - 3xN np matrix in tl coord frame. # @param a - motion of the segway (in the ms frame) # @return vecs_ts def tlRts(vecs_ts, a): vecs_ms = mcf.mecanumTglobal(mcf.globalTtorso(vecs_ts, True), True) vecs_ml = tr.Rz(a) * vecs_ms vecs_tl = mcf.torsoTglobal(mcf.globalTmecanum(vecs_ml, True), True) return vecs_tl def pts_within_dist(p,pts,min_dist,max_dist): v = p-pts d_arr = ut.norm(v).A1 idxs = np.where(np.all(np.row_stack((d_arr<max_dist,d_arr>min_dist)),axis=0)) pts_within = pts[:,idxs[0]] return pts_within ## apologies for the poor name. computes the translation of the torso # frame that move the eq pt away from closest boundary and rotate such # that local x axis is perp to mechanism returns 2x1 np matrix, angle def segway_motion_repulse(curr_pos_tl, eq_pt_tl,bndry, all_pts): bndry_dist_eq = dist_from_boundary(eq_pt_tl,bndry,all_pts) # signed bndry_dist_ee = dist_from_boundary(curr_pos_tl,bndry,all_pts) # signed if bndry_dist_ee < bndry_dist_eq: p = curr_pos_tl[0:2,:] bndry_dist = bndry_dist_ee else: p = eq_pt_tl[0:2,:] bndry_dist = bndry_dist_eq # p = eq_pt_tl[0:2,:] pts_close = pts_within_dist(p,bndry,0.002,0.07) v = p-pts_close d_arr = ut.norm(v).A1 v = v/d_arr v = v/d_arr # inverse distance weight resultant = v.sum(1) res_norm = np.linalg.norm(resultant) resultant = resultant/res_norm tvec = -resultant if bndry_dist < 0.: tvec = -tvec # eq pt was outside workspace polygon. if abs(bndry_dist)<0.01 or res_norm<0.01: # internal external test fails so falling back on # going to mean. m = all_pts.mean(1) tvec = m-p tvec = -tvec/np.linalg.norm(tvec) dist_move = 0. if bndry_dist > 0.05: dist_move = 0. else: dist_move = 1. tvec = tvec*dist_move # tvec is either a unit vec or zero vec. return tvec if __name__ == '__main__': #d = ut.load_pickle('workspace_dict_2009Sep03_221107.pkl') d = ut.load_pickle('../../pkls/workspace_dict_2009Sep05_200116.pkl') z = -0.23 k = d.keys() k_idx = np.argmin(np.abs(np.array(k)-z)) pts = d[k[k_idx]] # visualize_boundary() for kk in k: pts = d[kk] bpts = compute_boundary(pts) cx,cy = 0.7,-0.6 rad = 0.4 # pts_in = point_contained(cx,cy,0.,rad,pts, # start_angle=math.radians(140), # end_angle=math.radians(190)) mpu.figure() mpu.plot_yx(pts[1,:].A1,pts[0,:].A1,linewidth=0) mpu.plot_yx(bpts[1,:].A1,bpts[0,:].A1,linewidth=0,color='y') # mpu.plot_yx(pts_in[1,:].A1,pts_in[0,:].A1,linewidth=0,color='g') # mpu.plot_yx([cy],[cx],linewidth=0,color='r') mpu.show() ### apologies for the poor name. computes the translation and rotation ## of the torso frame that move the eq pt away from closest boundary ## and rotate such that local x axis is perp to mechanism ## returns 2x1 np matrix, angle #def segway_motion_repulse(curr_pos_tl,cx_tl,cy_tl,cy_ts,start_pos_ts, # eq_pt_tl,bndry): # vec_bndry = vec_from_boundary(eq_pt_tl,bndry) # dist_boundary = np.linalg.norm(vec_bndry) # vec_bndry = vec_bndry/dist_boundary # # radial_vec_tl = curr_pos_tl[0:2]-np.matrix([cx_tl,cy_tl]).T # radial_angle = math.atan2(radial_vec_tl[1,0],radial_vec_tl[0,0]) # if cy_ts<start_pos_ts[1,0]: # err = radial_angle-math.pi/2 # else: # err = radial_angle +math.pi/2 # # a_torso = err # dist_move = max(0.15-dist_boundary,0.) # if dist_move < 0.04: # dist_move = 0. # hook_translation_tl = -vec_bndry*dist_move # ## print 'vec_bndry:',vec_bndry.A1.tolist() ## print 'dist_boundary:',dist_boundary # # return hook_translation_tl,a_torso
[ [ 1, 0, 0.0794, 0.0025, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0819, 0.0025, 0, 0.66, 0.037, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0893, 0.0025, 0, 0.6...
[ "import roslib", "roslib.load_manifest('equilibrium_point_control')", "import numpy as np, math", "import scipy.optimize as so", "import scipy.ndimage as ni", "import matplotlib_util.util as mpu", "import hrl_lib.util as ut", "import hrl_lib.transforms as tr", "import hrl_hokuyo.hokuyo_processing as...
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('2010_icra_epc_pull') import rospy from 2010_icra_epc_pull.msg import MechanismKinematicsRot from geometry_msgs.msg import Point32 import arm_trajectories as at from threading import RLock import numpy as np import time ## # fit circle to the trajectory, publish the computed kinematics. # @param cartesian_pts_list - list of 3-tuples. trajectory of the mechanism # @param pbshr - publisher for the MechanismKinematics message # @param lock - to make list operations thread safe. (there is a callback too.) def circle_estimator(cartesian_pts_list, pbshr, lock): lock.acquire() n_pts = len(cartesian_pts_list) pts_2d = (np.matrix(cartesian_pts_list).T)[0:2,:] lock.release() if n_pts<2: time.sleep(0.1) #pbshr.publish(mk) # don't publish anything. return st = pts_2d[:,0] now = pts_2d[:,-1] mk = MechanismKinematicsRot() mk.cx = 0.5 mk.cy = -3.5 mk.cz = cartesian_pts_list[0][2] mk.rad = 10. dist_moved = np.linalg.norm(st-now) if dist_moved<=0.1: reject_pts_num = n_pts else: reject_pts_num = 1 if dist_moved<=0.15: time.sleep(0.1) pbshr.publish(mk) return # pts_2d = (np.matrix(cartesian_pts_list).T)[0:2,:] pts_2d = pts_2d[:,reject_pts_num:] rad = 1.0 #start_pos = np.matrix(cartesian_pts_list[0]).T start_pos = st rad,cx,cy = at.fit_circle(rad, start_pos[0,0], start_pos[1,0]-rad, pts_2d, method='fmin_bfgs', verbose=False) mk.cx = cx mk.cy = cy mk.rad = rad pbshr.publish(mk) # append the point to the trajectory def trajectory_cb(pt32, tup): cp_list, lock = tup lock.acquire() cp_list.append([pt32.x, pt32.y, pt32.z]) lock.release() if __name__ == '__main__': cartesian_points_list = [] lock = RLock() rospy.init_node('kinematics_estimator_least_sq') mech_kin_pub = rospy.Publisher('mechanism_kinematics_rot', MechanismKinematicsRot) rospy.Subscriber('mechanism_trajectory', Point32, trajectory_cb, (cartesian_points_list, lock)) print 'Begin' while not rospy.is_shutdown(): circle_estimator(cartesian_points_list, mech_kin_pub, lock) time.sleep(0.01) print 'End'
[ [ 1, 0, 0.2735, 0.0085, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.2735, 0.0085, 0, 0.66, 0.1, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.2821, 0.0085, 0, 0.66,...
[ "import roslib; roslib.load_manifest('2010_icra_epc_pull')", "import roslib; roslib.load_manifest('2010_icra_epc_pull')", "import rospy", "from geometry_msgs.msg import Point32", "import arm_trajectories as at", "from threading import RLock", "import numpy as np", "import time", "def circle_estimato...
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import sys,os sys.path.append(os.environ['HRLBASEPATH']+'/usr/advait/LPI') import cam_utm_lpi as cul import hrl_lib.util as ut import hrl_lib.transforms as tr import mekabot.coord_frames as mcf import math, numpy as np import util as uto import tilting_hokuyo.processing_3d as p3d import camera_config as cc ## returns selected location in global coord frame. # @param angle - angle at which to take image, and about which to take # a 3D scan. def select_location(c,thok,angle): thok.servo.move_angle(angle) cvim = c.get_frame() cvim = c.get_frame() cvim = c.get_frame() im_angle = thok.servo.read_angle() tilt_angles = (math.radians(-20)+angle,math.radians(30)+angle) pos_list,scan_list = thok.scan(tilt_angles,speed=math.radians(10)) m = p3d.generate_pointcloud(pos_list,scan_list,math.radians(-60), math.radians(60), 0.0,-0.055) pts = mcf.utmcam0Tglobal(mcf.globalTthok0(m),im_angle) cam_params = cc.camera_parameters['mekabotUTM'] fx = cam_params['focal_length_x_in_pixels'] fy = cam_params['focal_length_y_in_pixels'] cx,cy = cam_params['optical_center_x_in_pixels'],cam_params['optical_center_y_in_pixels'] cam_proj_mat = np.matrix([[fx, 0, cx], [0, fy, cy], [0, 0, 1]]) cvim,pts2d = cul.project_points_in_image(cvim,pts,cam_proj_mat) cp = cul.get_click_location(cvim) print 'Clicked location:', cp if cp == None: return None idx = cul.get_index(pts2d.T,cp) pt3d = pts[:,idx] pt_interest = mcf.globalTutmcam0(pt3d,im_angle) hl_thok0 = mcf.thok0Tglobal(pt_interest) l1,l2 = 0.0,-0.055 d = {} d['pt'] = hl_thok0 d['pos_list'],d['scan_list'] = pos_list,scan_list d['l1'],d['l2'] = l1,l2 d['img'] = uto.cv2np(cvim) ut.save_pickle(d,'hook_plane_scan_'+ut.formatted_time()+'.pkl') return pt_interest if __name__ == '__main__': import camera import hokuyo.hokuyo_scan as hs import tilting_hokuyo.tilt_hokuyo_servo as ths hok = hs.Hokuyo('utm',0,flip=True,ros_init_node=True) thok = ths.tilt_hokuyo('/dev/robot/servo0',5,hok,l1=0.,l2=-0.055) cam = camera.Camera('mekabotUTM') for i in range(10): cam.get_frame() pt = select_location(cam,thok) print 'Selected location in global coordinates:', pt.A1.tolist()
[ [ 1, 0, 0.301, 0.0097, 0, 0.66, 0, 509, 0, 2, 0, 0, 509, 0, 0 ], [ 8, 0, 0.3107, 0.0097, 0, 0.66, 0.0909, 243, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.3204, 0.0097, 0, 0.6...
[ "import sys,os", "sys.path.append(os.environ['HRLBASEPATH']+'/usr/advait/LPI')", "import cam_utm_lpi as cul", "import hrl_lib.util as ut", "import hrl_lib.transforms as tr", "import mekabot.coord_frames as mcf", "import math, numpy as np", "import util as uto", "import tilting_hokuyo.processing_3d a...
#!/usr/bin/python import sys, os sys.path.append(os.environ['HRLBASEPATH']+'/src/libraries/') import cameras.dragonfly as dr import roslib; roslib.load_manifest('modeling_forces') import rospy import cv from collections import deque import time import math, numpy as np import glob import hrl_lib.util as ut import hrl_camera.ros_camera as rc def got_pose_cb(data, got_pose_dict): if len(data.objects) != 2: got_pose_dict['pose_fail'] = True else: got_pose_dict['pose_fail'] = False got_pose_dict['flag'] = True if __name__ == '__main__': import optparse p = optparse.OptionParser() p.add_option('-l', '--log', action='store_true', dest='log', help='log FT data') p.add_option('-p', '--pub', action='store_true', dest='pub', help='publish over ROS') p.add_option('-c', '--conv', action='store_true', dest='avi_to_pngs', help='convert avi to pngs') p.add_option('-b', '--bad', action='store_true', dest='bad', help='find the images on which checker tracking failed.') p.add_option('-d', '--dir', action='store', default='', type='string', dest='dir', help='directory with images') p.add_option('-i', '--images_only', action='store_true', dest='images_only', help='work with images (no pkl)') p.add_option('-s', '--single_im', action='store', default='', type='string', dest='single_fname', help='work with one image') opt, args = p.parse_args() camera_name = 'remote_head' if opt.pub: import cv from cv_bridge.cv_bridge import CvBridge, CvBridgeError from std_msgs.msg import String from std_msgs.msg import Empty from sensor_msgs.msg import Image from sensor_msgs.msg import CameraInfo from checkerboard_detector.msg import ObjectDetection rospy.init_node('publish_log_images', anonymous=True) if opt.single_fname != '': im_name_list = [opt.single_fname for i in range(10)] time_list = [time.time() for i in range(10)] elif opt.images_only: im_name_list = glob.glob(opt.dir+'/*.png') #im_name_list = glob.glob(opt.dir+'/*.jpg') im_name_list.sort() time_list = [1 for i in range(len(im_name_list))] else: l = glob.glob(opt.dir+'/handheld_pull_log*.pkl') if l == []: raise RuntimeError('%s does not have a handheld_pull_log'%opt.dir) pkl_name = l[0] d = ut.load_pickle(pkl_name) im_name_list = glob.glob(opt.dir+'/0*.png') im_name_list.sort() time_list = d['time_list'] import camera_config as cc cp = cc.camera_parameters[camera_name] m = np.array([ [ cp['focal_length_x_in_pixels'], 0., cp['optical_center_x_in_pixels'], 0. ], [ 0., cp['focal_length_y_in_pixels'], cp['optical_center_y_in_pixels'], 0. ], [ 0., 0., 1., 0.] ]) intrinsic_list = [m[0,0], m[0,1], m[0,2], 0.0, m[1,0], m[1,1], m[1,2], 0.0, m[2,0], m[2,1], m[2,2], 0.0] topic_name = 'cvcamera_' + camera_name image_pub = rospy.Publisher(topic_name, Image) config_pub = rospy.Publisher(topic_name+'_info', CameraInfo) ch_pub = rospy.Publisher('/checker_to_poses/trigger', Empty) time.sleep(0.5) bridge = CvBridge() got_pose_dict = {'flag': False, 'pose_fail': False} topic_name_cb = '/checkerdetector/ObjectDetection' rospy.Subscriber(topic_name_cb, ObjectDetection, got_pose_cb, got_pose_dict) failed_im_list = [] # list of filenames on which checkboard detection failed. n_images = len(im_name_list) for i in range(n_images): name = im_name_list[i] cv_im = cv.LoadImage(name) rosimage = bridge.cv_to_imgmsg(cv_im, "bgr8") rosimage.header.stamp = rospy.Time.from_sec(time_list[i]) image_pub.publish(rosimage) config_pub.publish(CameraInfo(P=intrinsic_list)) t_st = time.time() while got_pose_dict['flag'] == False: time.sleep(0.5) if (time.time()-t_st) > 10.: break if got_pose_dict['pose_fail'] == True: failed_im_list.append(name) time.sleep(0.5) got_pose_dict['flag'] = False got_pose_dict['pose_fail'] = False if rospy.is_shutdown(): break print 'Number of images:', n_images ch_pub.publish() # send trigger to the ft logger. ut.save_pickle(failed_im_list, 'checker_fail_list.pkl') if opt.log: from opencv.cv import * from opencv.highgui import * from std_msgs.msg import Empty rospy.init_node('image logger', anonymous=True) ft_pub = rospy.Publisher('/ftlogger/trigger', Empty) cam = dr.dragonfly2(camera_name) cam.set_frame_rate(30) cam.set_brightness(0, 651, 0, 65) for i in range(10): im = cam.get_frame_debayered() # undistorting slows down frame rate im_list = deque() time_list = [] cvNamedWindow('Image Logging', CV_WINDOW_AUTOSIZE) print 'Started the loop.' print 'Hit a to start logging, ESC to exit and save pkl' log_images = False while not rospy.is_shutdown(): kp = cvWaitKey(1) if (type(kp) == str and kp == '\x1b') or (type(kp) != str and kp & 255 == 27): # ESC then exit. t1 = time.time() ft_pub.publish() # send trigger to the ft logger. break if (type(kp) == str and kp == 'a') or (type(kp) != str and kp & 255 == 97): # a to start logging. log_images = True t0 = time.time() ft_pub.publish() # send trigger to the ft logger. print 'started logging' im = cam.get_frame_debayered() # undistorting slows down frame rate if log_images: time_list.append(time.time()) im_list.append(cvCloneImage(im)) print 'frame rate:', len(time_list)/(t1-t0) print 'before saving the pkl' d = {} t_string = ut.formatted_time() video_name = 'mechanism_video_'+t_string+'.avi' vwr = cvCreateVideoWriter(video_name, CV_FOURCC('I','4','2','0'), 30, cvGetSize(im_list[0]), True) t0 = time.time() im_name_list = [] time_stamp = ut.formatted_time() for im in im_list: cvWriteFrame(vwr, im) time.sleep(.01) #Important to keep force torque server #from restarting t1 = time.time() print 'disk writing rate:', len(time_list)/(t1-t0) d['time_list'] = time_list d['video_name'] = video_name fname = 'handheld_pull_log_' + t_string + '.pkl' ut.save_pickle(d, fname) print 'Done saving the pkl' if opt.avi_to_pngs: from opencv.cv import * from opencv.highgui import * import util import camera_config as cc cp = cc.camera_parameters[camera_name] size = (int(cp['calibration_image_width']), int(cp['calibration_image_height'])) color = cp['color'] intrinsic_cvmat = cvCreateMat(3,3,cv.CV_32FC1) distortion_cvmat = cvCreateMat(1,4,cv.CV_32FC1) imat_np = np.array([[cp['focal_length_x_in_pixels'],0, cp['optical_center_x_in_pixels']], [0,cp['focal_length_y_in_pixels'], cp['optical_center_y_in_pixels']], [0,0,1]]) intrinsic_cvmat = util.numpymat2cvmat(imat_np) dmat_np = np.array([[cp['lens_distortion_radial_1'], cp['lens_distortion_radial_2'], cp['lens_distortion_tangential_1'], cp['lens_distortion_tangential_2']]]) distortion_cvmat = util.numpymat2cvmat(dmat_np) undistort_mapx = cvCreateImage(size, IPL_DEPTH_32F, 1) undistort_mapy = cvCreateImage(size, IPL_DEPTH_32F, 1) cvInitUndistortMap(intrinsic_cvmat, distortion_cvmat, undistort_mapx, undistort_mapy) if color == True: undistort_image = cvCreateImage(size, IPL_DEPTH_8U, 3) else: undistort_image = cvCreateImage(size, IPL_DEPTH_8U, 1) #pkl_name = glob.glob(opt.dir+'/handheld_pull_log*.pkl')[0] #d = ut.load_pickle(pkl_name) #video_name = opt.dir+'/'+d['video_name'] #time_list = d['time_list'] video_name = glob.glob(opt.dir + 'mechanism_video*.avi')[0] cap = cvCreateFileCapture(video_name) #for i in range(len(time_list)): i = 0 while True: cvim = cvQueryFrame(cap) if cvim == None: break cvFlip(cvim, cvim) # undistort the image cvRemap(cvim, undistort_image, undistort_mapx, undistort_mapy, CV_INTER_LINEAR, cvScalarAll(0)) nm = opt.dir+'/%05d.png'%i print 'Saving', nm cvSaveImage(nm, undistort_image) i += 1 if opt.bad: import cv l = ut.load_pickle(opt.dir+'/checker_fail_list.pkl') display = False if display: wnd = 'Checker Fail Images' cv.NamedWindow(wnd, cv.CV_WINDOW_AUTOSIZE) else: save_dir = opt.dir+'/checker_fail/' os.system('mkdir %s'%save_dir) for nm in l: name = opt.dir+'/'+nm cv_im = cv.LoadImage(name) if display: cv.ShowImage(wnd, cv_im) cv.WaitKey(0) else: save_dir = os.path.normpath(save_dir) # print 'save_dir:', save_dir file_name = '_'.join(save_dir.split('/')) + '_%s'%os.path.normpath(nm) print 'file_name:', file_name cv.SaveImage(save_dir + '/' + file_name, cv_im)
[ [ 1, 0, 0.0103, 0.0034, 0, 0.66, 0, 509, 0, 2, 0, 0, 509, 0, 0 ], [ 8, 0, 0.0137, 0.0034, 0, 0.66, 0.0714, 243, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0172, 0.0034, 0, 0....
[ "import sys, os", "sys.path.append(os.environ['HRLBASEPATH']+'/src/libraries/')", "import cameras.dragonfly as dr", "import roslib; roslib.load_manifest('modeling_forces')", "import roslib; roslib.load_manifest('modeling_forces')", "import rospy", "import cv", "from collections import deque", "impor...
# # Assumes that all the logs are in one folder with the file names # giving the different mechanisms, trial number, open or close etc. # import commands import os import os.path as pt import glob import math, numpy as np import scipy.signal as ss import scipy.cluster as clus import roslib; roslib.load_manifest('modeling_forces') import modeling_forces.mf_common as mfc import matplotlib_util.util as mpu import hrl_lib.util as ut import scipy.stats as st import pylab as pb #import mdp class pca_plot_gui(): def __init__(self, legend_list, mech_vec_list, proj_mat, dir_list, mn): self.legend_list = legend_list self.mech_vec_list = mech_vec_list self.proj_mat = proj_mat self.dir_list = dir_list self.mn = mn def pick_cb(self, event): if 'shift' != event.key: return selected = np.matrix([event.xdata, event.ydata]).T # print 'selected', selected.A1 min_list = [] for i, v in enumerate(self.mech_vec_list): p = self.proj_mat[:, 0:2].T * (v-self.mn) min_list.append(np.min(ut.norm(p-selected))) mech_idx = np.argmin(min_list) print 'Selected mechanism was:', self.legend_list[mech_idx] plot_tangential_force(self.dir_list[mech_idx], all_trials = True, open = True) # mpu.figure() # n_dim = 7 # reconstruction = self.proj_mat[:, 0:n_dim] * (self.proj_mat[:, 0:n_dim].T * (self.mech_vec_list[mech_idx] - self.mn)) + self.mn # di = extract_pkls(self.dir_list[mech_idx]) # nm = get_mech_name(self.dir_list[mech_idx]) # for i in range(reconstruction.shape[1]): # mpu.plot_yx(reconstruction[:,i].A1, color=mpu.random_color(), # plot_title=nm+': %d components'%n_dim) # mpu.legend() # mpu.figure() # plot_velocity(self.dir_list[mech_idx]) mpu.show() ## # list all the mechanisms in dir_list and allow user to type the # numbers of the desired mechanisms. # @return list of paths to the selected mechanisms. def input_mechanism_list(dir_list): mech_list = [] for i, m in enumerate(dir_list): t = m.split('/') mech = t[-1] if mech == '': mech = t[-2] mech_list.append(mech) print '%d. %s'%(i, mech) print '' print 'Enter mechanism numbers that you want to plot' s = raw_input() num_list = map(int, s.split(' ')) chosen_list = [] for n in num_list: chosen_list.append(dir_list[n]) return chosen_list ## # remove pkls in which the forces are unreasonable large or small def clean_data_forces(dir): l = commands.getoutput('find %s/ -name "*mechanism_trajectories*.pkl"'%dir).splitlines() for pkl in l: cmd1 = 'rm -f %s'%pkl cmd2 = 'svn rm %s'%pkl d = ut.load_pickle(pkl) radial_mech = d['force_rad_list'] tangential_mech = d['force_tan_list'] if len(radial_mech) == 0 or len(tangential_mech) == 0: os.system(cmd1) os.system(cmd2) continue max_force = max(np.max(np.abs(radial_mech)), np.max(np.abs(tangential_mech))) if max_force > 120.: os.system(cmd1) os.system(cmd2) n_points = len(radial_mech) if n_points < 50: os.system(cmd1) os.system(cmd2) if d.has_key('radius'): r = d['radius'] if r != -1: ang = np.degrees(d['mechanism_x']) if np.max(ang) < 20.: os.system(cmd1) os.system(cmd2) if d.has_key('time_list'): t_l = d['time_list'] time_diff_l = np.array(t_l[1:]) - np.array(t_l[0:-1]) if len(time_diff_l) == 0: os.system(cmd1) os.system(cmd2) continue max_time_diff = np.max(time_diff_l) if max_time_diff > 0.3: print 'max time difference between consec readings:', max_time_diff os.system(cmd1) os.system(cmd2) ## # @param dir_name - directory containing the pkls. # @param all_trials - plot force for all the trials. # @param open - Boolean (open or close trial) # @param filter_speed - mech vel above this will be ignored. for # ROTARY joints only, radians/sec def plot_tangential_force(dir_name, all_trials, open = True, filter_speed=math.radians(100)): mpu.set_figure_size(4.,4.) fig1 = mpu.figure() # fig2 = mpu.figure() # fig3 = mpu.figure() if open: trial = 'open' else: trial = 'close' d = extract_pkls(dir_name, open) mech_name = get_mech_name(dir_name) traj_vel_list = [] for i,ftan_l in enumerate(d['ftan_l_l']): mech_x = d['mechx_l_l'][i] if d['typ'] == 'rotary': max_angle = math.radians(30) type = 'rotary' else: max_angle = 0.3 type = 'prismatic' traj_vel = compute_average_velocity(mech_x, d['time_l_l'][i], max_angle, type) print 'traj_vel:', traj_vel if traj_vel == -1: continue traj_vel_list.append(traj_vel) #vel_color_list = ['r', 'g', 'b', 'y', 'k'] #vel_color_list = ['#000000', '#A0A0A0', '#D0D0D0', '#E0E0E0', '#F0F0F0'] #vel_color_list = ['#000000', '#00A0A0', '#00D0D0', '#00E0E0', '#00F0F0'] #vel_color_list = [ '#%02X%02X%02X'%(r,g,b) for (r,g,b) in [(95, 132, 53), (81, 193, 79), (28, 240, 100), (196, 251, 100)]] vel_color_list = [ '#%02X%02X%02X'%(r,g,b) for (r,g,b) in [(95, 132, 53), (28, 240, 100), (196, 251, 100)]] traj_vel_sorted = np.sort(traj_vel_list).tolist() sorted_color_list = [] sorted_scatter_list = [] i = 0 v_prev = traj_vel_sorted[0] legend_list = [] if d['typ'] == 'rotary': l = '%.1f'%math.degrees(v_prev) thresh = math.radians(5) else: l = '%.02f'%(v_prev) thresh = 0.05 t_v = v_prev v_threshold = 1000. for j,v in enumerate(traj_vel_sorted): if (v - v_prev) > thresh: if d['typ'] == 'rotary': l = l + ' to %.1f deg/sec'%math.degrees(t_v) else: l = l + ' to %.1f m/s'%t_v legend_list.append(l) if d['typ'] == 'rotary': l = '%.1f'%math.degrees(v) else: l = '%.02f'%(v) i += 1 if i >= 2: v_threshold = min(v_threshold, v) if d['typ'] == 'rotary': print 'v_threshold:', math.degrees(v_threshold) else: print 'v_threshold:', v_threshold if i == len(vel_color_list): i -= 1 v_prev = v else: legend_list.append('__nolegend__') t_v = v sorted_color_list.append(vel_color_list[i]) if d['typ'] == 'rotary': l = l + ' to %.1f deg/sec'%math.degrees(t_v) else: l = l + ' to %.1f m/s'%t_v legend_list.append(l) legend_list = legend_list[1:] giant_list = [] mpu.set_figure_size(3.,3.) for i,ftan_l in enumerate(d['ftan_l_l']): mech_x = d['mechx_l_l'][i] trial_num = str(d['trial_num_l'][i]) color = None scatter_size = None traj_vel = compute_average_velocity(mech_x, d['time_l_l'][i], max_angle, d['typ']) if traj_vel == -1: continue if traj_vel >= v_threshold: continue color = sorted_color_list[traj_vel_sorted.index(traj_vel)] legend = legend_list[traj_vel_sorted.index(traj_vel)] if d['typ'] == 'rotary': #traj_vel = compute_trajectory_velocity(mech_x,d['time_l_l'][i],1) #if traj_vel >= filter_speed: # continue mech_x_degrees = np.degrees(mech_x) xlabel = 'angle (degrees)' ylabel = '$f_{tan}$ (N)' else: mech_x_degrees = mech_x xlabel = 'distance (meters)' ylabel = 'Opening force (N)' # n_skip = 65 # print '>>>>>>>>>>>>> WARNING BEGIN <<<<<<<<<<<<<<<<<' # print 'not plotting the last ', n_skip, 'data points for drawers' # print '>>>>>>>>>>>>> WARNING END <<<<<<<<<<<<<<<<<' n_skip = 1 mech_x_degrees = mech_x_degrees[:-n_skip] ftan_l = ftan_l[:-n_skip] mpu.figure(fig1.number) #color, scatter_size = None, None scatter_size = None if color == None: color = mpu.random_color() if scatter_size == None: scatter_size = 1 giant_list.append((traj_vel, ftan_l, mech_x_degrees, color, legend, xlabel, ylabel, trial_num)) giant_list_sorted = reversed(sorted(giant_list)) for traj_vel, ftan_l, mech_x_degrees, color, legend, xlabel, ylabel, trial_num in giant_list_sorted: mpu.plot_yx(ftan_l, mech_x_degrees, axis=None, plot_title= '\huge{%s: %s}'%(mech_name, trial), xlabel=xlabel, ylabel = ylabel, color = color, #scatter_size = 5, linewidth = 1, label=trial_num) scatter_size = scatter_size, linewidth = 1, label=legend) print '>>>>>>>>>> number of trials <<<<<<<<<<<', len(giant_list) mpu.figure(fig1.number) # mpu.legend(display_mode='less_space') def radial_tangential_ratio(dir_name): d = extract_pkls(dir_name, open=True) nm = get_mech_name(dir_name) frad_ll = d['frad_l_l'] mechx_ll = d['mechx_l_l'] mpu.figure() for i,ftan_l in enumerate(d['ftan_l_l']): frad_l = frad_ll[i] rad_arr = np.array(np.abs(frad_l)) tan_arr = np.array(np.abs(ftan_l)) idxs = np.where(np.logical_and(rad_arr > 0.1, tan_arr > 0.1)) ratio = np.divide(rad_arr[idxs], tan_arr[idxs]) mpu.plot_yx(ratio, np.degrees(np.array(mechx_ll[i])[idxs]), color = mpu.random_color(), plot_title=nm, ylabel='radial/tangential', xlabel='Angle (degrees)') ## # get mechanism name from the directory name. def get_mech_name(d): t = d.split('/') mech_name = t[-1] if mech_name == '': mech_name = t[-2] return mech_name ## # get all the information from all the pkls in one directory. # ASSUMES - all pkls are of the same mechanism (same radius, type ...) # @param open - extract info for opening trials def extract_pkls(d, open=True, quiet = False, ignore_moment_list=False): if open: trial = 'open' else: trial = 'close' l = glob.glob(d+'/*'+trial+'*mechanism_trajectories*.pkl') l.sort() ftan_l_l, frad_l_l, mechx_l_l = [], [], [] time_l_l, trial_num_l = [], [] moment_l_l = [] typ, rad = None, None for p in l: d = ut.load_pickle(p) if d.has_key('radius'): rad = d['radius'] else: rad = -1 # if quiet == False: # print p, 'does not have radius' # return None if rad != -1: #moment_l_l.append((np.array(ftan_l_l[-1]) * rad).tolist()) if d.has_key('moment_list'): moment_l_l.append(d['moment_list']) else: # if quiet == False: # print p, 'does not have moment_list' # continue moment_l_l.append([0 for i in range(len(d['force_rad_list']))]) trial_num = p.split('_')[-5] trial_num_l.append(int(trial_num)) frad_l_l.append(d['force_rad_list']) ftan_l_l.append(d['force_tan_list']) if d.has_key('mech_type'): typ = d['mech_type'] else: if quiet == False: print p, 'does not have mech_typ' return None mechx_l_l.append(d['mechanism_x']) if d.has_key('time_list'): t_l = d['time_list'] else: t_l = [0.03*i for i in range(len(ftan_l_l))] #if quiet == False: # print p, 'does not have time_list' #return None time_l_l.append((np.array(t_l)-t_l[0]).tolist()) r = {} r['ftan_l_l'] = ftan_l_l r['frad_l_l'] = frad_l_l r['mechx_l_l'] = mechx_l_l r['typ'] = typ r['rad'] = rad r['time_l_l'] = time_l_l r['trial_num_l'] = trial_num_l r['moment_l_l'] = moment_l_l return r ## # take max force magnitude within a bin size # @param bin_size - depends on the units of poses_list # @param fn - function to apply to the binned force values (e.g. max, min) # @param ignore_empty - if True then empty bins are ignored. Else the # value of an empty bin is set to None. # @param max_pose - maximum value of pose to use if None then derived # from poses_list # @param empty_value - what to fill in an empty bin (None, np.nan etc.) def bin(poses_list, ftan_list, bin_size, fn, ignore_empty, max_pose=None, empty_value = None): if max_pose == None: max_dist = max(poses_list) else: max_dist = max_pose poses_array = np.array(poses_list) binned_poses_array = np.arange(0., max_dist, bin_size) binned_force_list = [] binned_poses_list = [] ftan_array = np.array(ftan_list) for i in range(binned_poses_array.shape[0]-1): idxs = np.where(np.logical_and(poses_array>=binned_poses_array[i], poses_array<binned_poses_array[i+1])) if idxs[0].shape[0] != 0: binned_poses_list.append(binned_poses_array[i]) binned_force_list.append(fn(ftan_array[idxs])) elif ignore_empty == False: binned_poses_list.append(binned_poses_array[i]) binned_force_list.append(empty_value) return binned_poses_list, binned_force_list ## # makes a scatter plot with the radius along the x axis and the max # force along the y-axis. different color for each mechanism # @param dir_name - directory containing the pkls. def max_force_radius_scatter(dir_name_list, open=True): mpu.figure() if open: trial = 'Opening' else: trial = 'Closing' for d in dir_name_list: nm = get_mech_name(d) print '>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>' print 'MECHANISM:', nm print '>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>' ftan_l_l, frad_l_l, mechx_l_l, typ, rad = extract_pkls(d, open) fl, rl = [], [] for n in range(len(ftan_l_l)): fmax = np.max(np.abs(ftan_l_l[n])) print 'fmax:', fmax fl.append(fmax) rl.append(rad) print 'Aloha' print 'len(fl)', len(fl) print 'len(rl)', len(rl) #mpu.plot_yx(fl, rl) plot_title = 'Scatter plot for %s trials'%trial mpu.plot_yx(fl, rl, color=mpu.random_color(), label=nm, axis=None, linewidth=0, xlabel='Radius (m)', ylabel='Maximum Force (N)', plot_title=plot_title) mpu.legend() # use all the values of all the bins for all the trials. # @param plot_type - 'tangential', 'magnitude', 'radial' def errorbar_one_mechanism(dir_name, open=True, new_figure=True, filter_speed = math.radians(100), plot_type='tangential', color=None, label = None): if new_figure: mpu.figure() nm = get_mech_name(dir_name) d = extract_pkls(dir_name, open) ftan_l_l = d['ftan_l_l'] frad_l_l = d['frad_l_l'] mechx_l_l = d['mechx_l_l'] time_l_l = d['time_l_l'] typ = d['typ'] rad = d['rad'] fn = list binned_mechx_l = [] binned_ftan_ll = [] use_trials_list = [] if plot_type == 'radial': force_l_l = frad_l_l if plot_type == 'tangential': force_l_l = ftan_l_l if plot_type == 'magnitude': force_l_l = [] for ta, ra in zip(ftan_l_l, frad_l_l): force_l_l.append(ut.norm(np.matrix([ta, ra])).A1.tolist()) n_trials = len(force_l_l) for i in range(n_trials): if typ == 'rotary': traj_vel = compute_trajectory_velocity(mechx_l_l[i], time_l_l[i], 1) if traj_vel >= filter_speed: continue t, f = bin(mechx_l_l[i], force_l_l[i], math.radians(1.), fn, ignore_empty=False, max_pose=math.radians(60)) if typ == 'prismatic': t, f = bin(mechx_l_l[i], force_l_l[i], 0.01, fn, ignore_empty=False, max_pose=0.5) if len(t) > len(binned_mechx_l): binned_mechx_l = t binned_ftan_ll.append(f) use_trials_list.append(i) n_trials = len(binned_ftan_ll) n_bins = len(binned_mechx_l) force_list_combined = [[] for i in binned_mechx_l] for i in range(n_trials): force_l = binned_ftan_ll[i] for j,p in enumerate(binned_mechx_l): if force_l[j] != None: if open: if j < 5: force_list_combined[j].append(max(force_l[j])) continue else: if (n_trials-j) < 5: force_list_combined[j].append(min(force_l[j])) continue force_list_combined[j] += force_l[j] plot_mechx_l = [] mean_l, std_l = [], [] for i,p in enumerate(binned_mechx_l): f_l = force_list_combined[i] if len(f_l) == 0: continue plot_mechx_l.append(p) mean_l.append(np.mean(f_l)) std_l.append(np.std(f_l)) if open: trial = 'Open' else: trial = 'Close' n_sigma = 1 if typ == 'rotary': x_l = np.degrees(plot_mechx_l) xlabel='\huge{Angle (degrees)}' else: x_l = plot_mechx_l xlabel='Distance (m)' std_arr = np.array(std_l) * n_sigma if color == None: color = mpu.random_color() if label == None: label= nm+' '+plot_type mpu.plot_errorbar_yx(mean_l, std_arr, x_l, linewidth=1, color=color, plot_title='\huge{Mean \& %d$\sigma$}'%(n_sigma), xlabel=xlabel, label=label, ylabel='\huge{Force (N)}') mpu.legend() # take the max of each bin. def errorbar_one_mechanism_max(dir_name, open=True, filter_speed=math.radians(100.)): # mpu.figure() nm = get_mech_name(dir_name) d = extract_pkls(dir_name, open) ftan_l_l = d['ftan_l_l'] frad_l_l = d['frad_l_l'] mechx_l_l = d['mechx_l_l'] time_l_l = d['time_l_l'] typ = d['typ'] rad = d['rad'] fn = max if open == False: fn = min binned_mechx_l = [] binned_ftan_ll = [] use_trials_list = [] n_trials = len(ftan_l_l) for i in range(n_trials): if typ == 'rotary': traj_vel = compute_trajectory_velocity(mechx_l_l[i], time_l_l[i], 1) if traj_vel >= filter_speed: continue t, f = bin(mechx_l_l[i], ftan_l_l[i], math.radians(1.), fn, ignore_empty=False, max_pose=math.radians(60)) if typ == 'prismatic': t, f = bin(mechx_l_l[i], ftan_l_l[i], 0.01, fn, ignore_empty=False, max_pose=0.5) if len(t) > len(binned_mechx_l): binned_mechx_l = t binned_ftan_ll.append(f) use_trials_list.append(i) binned_ftan_arr = np.array(binned_ftan_ll) plot_mechx_l = [] mean_l, std_l = [], [] for i,p in enumerate(binned_mechx_l): f_l = [] for j in range(len(use_trials_list)): if binned_ftan_arr[j,i] != None: f_l.append(binned_ftan_arr[j,i]) if len(f_l) == 0: continue plot_mechx_l.append(p) mean_l.append(np.mean(f_l)) std_l.append(np.std(f_l)) xlabel = 'Angle (degrees)' if open: trial = 'Open' else: trial = 'Close' n_sigma = 1 std_arr = np.array(std_l) * n_sigma mpu.plot_errorbar_yx(mean_l, std_arr, np.degrees(plot_mechx_l), linewidth=1, plot_title=nm+': '+trial, xlabel='Angle (degrees)', label='Mean \& %d$\sigma$'%(n_sigma), ylabel='Tangential Force (N)', color='y') mpu.legend() def plot_opening_distances_drawers(dir_name_list): mpu.figure() for d in dir_name_list: nm = get_mech_name(d) ftan_l_l, frad_l_l, mechx_l_l, typ, rad = extract_pkls(d) if rad != -1: # ignoring the cabinet doors. continue # print 'Aloha' # import pdb; pdb.set_trace() dist_opened_list = [] for x_l in mechx_l_l: dist_opened_list.append(x_l[-1] - x_l[0]) plot_title = 'Opening distance for drawers' mpu.plot_yx(dist_opened_list, color=mpu.random_color(), label=nm, axis=None, linewidth=0, xlabel='Nothing', ylabel='Distance opened', plot_title=plot_title) mpu.legend() def handle_height_histogram(dir_name_list, plot_title=''): mean_height_list = [] for d in dir_name_list: nm = get_mech_name(d) pkl = glob.glob(d+'/mechanism_calc_dict.pkl') if pkl == []: print 'Mechanism "%s" does not have a mechanism_calc_dict'%nm continue pkl = pkl[0] mech_calc_dict = ut.load_pickle(pkl) hb = mech_calc_dict['handle_bottom'] ht = mech_calc_dict['handle_top'] mean_height_list.append((hb+ht)/2.) #max_height = np.max(mean_height_list) max_height = 2.0 bin_width = 0.1 bins = np.arange(0.-bin_width/2., max_height+2*bin_width, bin_width) hist, bin_edges = np.histogram(np.array(mean_height_list), bins) mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=bin_width*0.8, plot_title=plot_title, xlabel='Height (meters)', ylabel='\# of mechanisms') def plot_handle_height(dir_name_list, plot_title): mpu.figure() for d in dir_name_list: nm = get_mech_name(d) pkl = glob.glob(d+'/mechanism_calc_dict.pkl') if pkl == []: print 'Mechanism "%s" does not have a mechanism_calc_dict'%nm continue pkl = pkl[0] mech_calc_dict = ut.load_pickle(pkl) hb = mech_calc_dict['handle_bottom'] ht = mech_calc_dict['handle_top'] di = extract_pkls(d, open=True) ftan_l_l = di['ftan_l_l'] frad_l_l = di['frad_l_l'] mechx_l_l = di['mechx_l_l'] time_l_l = di['time_l_l'] typ = di['typ'] rad = di['rad'] # ftan_l_l, frad_l_l, mechx_l_l, typ, rad = extract_pkls(d, # open=True) fl, hl = [], [] for n in range(len(ftan_l_l)): fmax = np.max(np.abs(ftan_l_l[n][0:-50])) fl.append(fmax) fl.append(fmax) hl.append(ht) hl.append(hb) mpu.plot_yx(hl, fl, color=mpu.random_color(), label=nm, axis=None, linewidth=0, xlabel='Max opening force', ylabel='Handle Height (m)', plot_title=plot_title) mpu.legend() def distance_of_handle_from_edges(): pass def plot_handle_height_no_office(): opt = commands.getoutput('cd aggregated_pkls_Feb11; ls --ignore=*HSI* --ignore=*HRL* --ignore=a.py') d_list = opt.splitlines() dir_list = [] for d in d_list: dir_list.append('aggregated_pkls_Feb11/'+d) plot_title = 'Only homes. Excluding Offices' plot_handle_height(dir_list[0:], plot_title) def plot_handle_height_no_fridge_no_freezer(): opt = commands.getoutput('cd aggregated_pkls_Feb11; ls --ignore=*refrigerator* --ignore=*freezer* --ignore=a.py') d_list = opt.splitlines() dir_list = [] for d in d_list: dir_list.append('aggregated_pkls_Feb11/'+d) plot_title = 'Excluding Refrigerators and Freezers' plot_handle_height(dir_list[0:], plot_title) ## # returns the median of the velocity. def compute_velocity(mech_x, time_list, smooth_window): x = np.array(mech_x) t = np.array(time_list) kin_info = {'disp_mech_coord_arr': np.array(mech_x), 'mech_time_arr': np.array(time_list)} vel_arr = mfc.velocity(kin_info, smooth_window) return vel_arr ## # mech_x must be in RADIANS. def compute_trajectory_velocity(mech_x, time_list, smooth_window): vel_arr = compute_velocity(mech_x, time_list, smooth_window) filt_vel_arr = vel_arr[np.where(vel_arr>math.radians(2.))] median_vel = np.median(filt_vel_arr) return median_vel ## # compute the average velocity = total angle / total time. def compute_average_velocity(mech_x, time_list, max_angle, type): reject_num = 20 if len(mech_x) < reject_num: return -1 mech_x = mech_x[reject_num:] time_list = time_list[reject_num:] if mech_x[-1] < max_angle: return -1 if type == 'rotary': start_angle = math.radians(1) elif type == 'prismatic': start_angle = 0.01 mech_x = np.array(mech_x) start_idx = np.where(mech_x > start_angle)[0][0] end_idx = np.where(mech_x > max_angle)[0][0] start_x = mech_x[start_idx] end_x = mech_x[end_idx] start_time = time_list[start_idx] end_time = time_list[end_idx] avg_vel = (end_x - start_x) / (end_time - start_time) return avg_vel def plot_velocity(dir_name): d = extract_pkls(dir_name, True) vel_fig = mpu.figure() acc_fig = mpu.figure() for i,time_list in enumerate(d['time_l_l']): mechx_l = d['mechx_l_l'][i] mechx_l, vel, acc, time_list = mfc.kinematic_params(mechx_l, time_list, 10) vel_arr = np.array(vel) acc_arr = np.array(acc) trial_num = d['trial_num_l'][i] xarr = np.array(mechx_l) idxs = np.where(np.logical_and(xarr < math.radians(20.), xarr > math.radians(1.))) color=mpu.random_color() mpu.figure(vel_fig.number) mpu.plot_yx(np.degrees(vel_arr[idxs]), np.degrees(xarr[idxs]), color=color, label='%d velocity'%trial_num, scatter_size=0) mpu.legend() mpu.figure(acc_fig.number) mpu.plot_yx(np.degrees(acc_arr[idxs]), np.degrees(xarr[idxs]), color=color, label='%d acc'%trial_num, scatter_size=0) mpu.legend() ## # l list of trials with which to correlate c1 # trial is a list of forces (each element is the max force or some # other representative value for a given angle) # lab_list - list of labels def correlate_trials(c1, l, lab_list): mpu.figure() x = 0 corr_list = [] x_l = [] for i,c2 in enumerate(l): res = ss.correlate(np.array(c1), np.array(c2), 'valid')[0] r1 = ss.correlate(np.array(c1), np.array(c1), 'valid')[0] r2 = ss.correlate(np.array(c2), np.array(c2), 'valid')[0] res = res/math.sqrt(r1*r2) # cross correlation coefficient http://www.staff.ncl.ac.uk/oliver.hinton/eee305/Chapter6.pdf if i == 0 or lab_list[i] == lab_list[i-1]: corr_list.append(res) x_l.append(x) else: mpu.plot_yx(corr_list, x_l, color=mpu.random_color(), label=lab_list[i-1], xlabel='Nothing', ylabel='Cross-Correlation Coefficient') corr_list = [] x_l = [] x += 1 mpu.plot_yx(corr_list, x_l, color=mpu.random_color(), label=lab_list[i-1]) mpu.legend() ## # plot errorbars showing 1 sigma for tangential component of the force # and the total magnitude of the force. (Trying to verify that what we # are capturing using our setup is consistent across people) def compare_tangential_total_magnitude(dir): mpu.figure() errorbar_one_mechanism(dir, open = True, filter_speed = math.radians(30), plot_type = 'magnitude', new_figure = False, color='y', label = '\huge{$\hat F_{normal}$}') errorbar_one_mechanism(dir, open = True, filter_speed = math.radians(30), plot_type = 'tangential', color='b', new_figure = False, label = '\huge{$||\hat F_{normal} + \hat F_{plane}||$}') def max_force_vs_velocity(dir): di = extract_pkls(dir, open) ftan_l_l = di['ftan_l_l'] frad_l_l = di['frad_l_l'] mechx_l_l = di['mechx_l_l'] time_l_l = di['time_l_l'] typ = di['typ'] rad = di['rad'] nm = get_mech_name(dir) # mpu.figure() color = mpu.random_color() mfl = [] tvl = [] for i in range(len(ftan_l_l)): xarr = np.array(mechx_l_l[i]) idxs = np.where(np.logical_and(xarr < math.radians(20.), xarr > math.radians(1.))) max_force = np.max(np.array(ftan_l_l[i])[idxs]) mechx_short = np.array(mechx_l_l[i])[idxs] time_short = np.array(time_l_l[i])[idxs] vel_arr = compute_velocity(mechx_l_l[i], time_l_l[i], 5) #vel_arr = compute_velocity(mechx_short[i], time_l_l[i], 5) vel_short = vel_arr[idxs] traj_vel = np.max(vel_short) #acc_arr = compute_velocity(vel_arr, time_l_l[i], 1) #traj_vel = np.max(acc_arr) #traj_vel = compute_trajectory_velocity(mechx_short, time_short, 1) mfl.append(max_force) tvl.append(traj_vel) mpu.plot_yx(mfl, tvl, color = color, xlabel = 'Trajectory vel', label = nm, ylabel = 'Max tangential force', linewidth=0) mpu.legend() def mechanism_radius_histogram(dir_list, color='b'): rad_list = [] for d in dir_list: nm = get_mech_name(d) pkl = glob.glob(d+'/mechanism_info.pkl') if pkl == []: print 'Mechanism "%s" does not have a mechanism_info_dict'%nm continue pkl = pkl[0] md = ut.load_pickle(pkl) if md['radius'] != -1: rad_list.append(md['radius']) max_radius = np.max(rad_list) print 'Rad list:', rad_list bin_width = 0.05 bins = np.arange(0.-bin_width/2., max_radius+2*bin_width, bin_width) hist, bin_edges = np.histogram(np.array(rad_list), bins) print 'Bin Edges:', bin_edges print 'Hist:', hist h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Radius(meters)', ylabel='\# of mechanisms', plot_title='Histogram of radii of rotary mechanisms', color=color) return h def make_vector(mechx, ftan_l, lim, bin_size): t, f = bin(mechx, ftan_l, bin_size, max, ignore_empty=False, max_pose=lim, empty_value = np.nan) f = np.array(f) t = np.array(t) clean_idx = np.where(np.logical_not(np.isnan(f))) miss_idx = np.where(np.isnan(f)) if len(miss_idx[0]) > 0: fclean = f[clean_idx] mechx_clean = t[clean_idx] mechx_miss = t[miss_idx] f_inter = mfc.interpolate_1d(mechx_clean, fclean, mechx_miss) f[np.where(np.isnan(f))] = f_inter #import pdb #pdb.set_trace() return np.matrix(f).T # this makes the vector unit norm before returning it. Currently, this # function is only used for PCA. def make_vector_mechanism(dir, use_moment = False): print '>>>>>>>>>>>>>>>>>>>>>>' print 'dir:', dir di = extract_pkls(dir) ftan_l_l = di['ftan_l_l'] frad_l_l = di['frad_l_l'] mechx_l_l = di['mechx_l_l'] time_l_l = di['time_l_l'] moment_l_l = di['moment_l_l'] typ = di['typ'] rad = di['rad'] n_trials = len(ftan_l_l) vec_list = [] tup_list = [] for i in range(n_trials): if typ == 'rotary': if use_moment: torque_l = moment_l_l[i] else: torque_l = ftan_l_l[i] if len(mechx_l_l[i]) < 30: continue v = make_vector(mechx_l_l[i], torque_l, lim = math.radians(50.), bin_size = math.radians(1)) max_angle = math.radians(30) if typ == 'prismatic': v = make_vector(mechx_l_l[i], ftan_l_l[i], lim = 0.25, bin_size = 0.01) traj_vel = compute_average_velocity(mechx_l_l[i], time_l_l[i], max_angle, typ) if traj_vel == -1: continue #v = v / np.linalg.norm(v) vec_list.append(v) tup_list.append((traj_vel,v)) if len(vec_list) <= 1: return None tup_list.sort() [vel_list, vec_list] = zip(*tup_list) v_prev = vel_list[0] t_v = v_prev thresh = math.radians(5) i = 0 ret_list = [] for j,v in enumerate(vel_list): if (v - v_prev) > thresh: i += 1 if i >= 2: break v_prev = v ret_list.append(vec_list[j]) print 'Number of trials:', len(ret_list) # selecting only the slowest three trials. # tup_list.sort() # if len(tup_list) > 3: # [acc_list, v_list] = zip(*tup_list) # return np.column_stack(v_list[0:3]) return np.column_stack(ret_list) ## # one of the figures for the paper. # showing that different classes have different clusters. def different_classes_rotary(dir_list): mech_vec_list = [] legend_list = [] for d in dir_list: di = extract_pkls(d) if di == None: continue if di.has_key('typ'): typ = di['typ'] if typ != 'rotary': #if typ != 'prismatic': continue else: continue v = make_vector_mechanism(d) if v == None: continue mech_vec_list.append(v) legend_list.append(get_mech_name(d)) all_vecs = np.column_stack(mech_vec_list) print '>>>>>>>>> all_vecs.shape <<<<<<<<<<<<<', all_vecs.shape U, s, _ = np.linalg.svd(np.cov(all_vecs)) mn = np.mean(all_vecs, 1).A1 mpu.set_figure_size(3.,3.) mpu.figure() proj_mat = U[:, 0:2] legend_made_list = [False, False, False] for i, v in enumerate(mech_vec_list): p = proj_mat.T * (v - np.matrix(mn).T) if np.any(p[0,:].A1<0): print 'First principal component < 0 for some trial of:', legend_list[i] color = mpu.random_color() if 'ree' in legend_list[i]: #color = 'g' color = '#66FF33' if legend_made_list[0] == False: label = 'Freezers' legend_made_list[0] = True else: label = '__nolegend__' print 'SHAPE:', p.shape print 'p:', p elif 'ge' in legend_list[i]: color = '#FF6633' #color = 'y' if legend_made_list[1] == False: label = 'Refrigerators' legend_made_list[1] = True else: label = '__nolegend__' else: #color = 'b' color = '#3366FF' if legend_made_list[2] == False: #label = '\\flushleft Cabinets, Spring \\\\*[-2pt] Loaded Doors' label = 'Cabinets' legend_made_list[2] = True else: label = '__nolegend__' mpu.pl.scatter(p[0,:].A1, p[1,:].A1, color = color, s = 15, label = label) mpu.pl.xlabel('First Principle Component') mpu.pl.ylabel('Second Principle Component') mpu.pl.axis('equal') mpu.pl.axhline(y=0., color = 'k', ls='--') mpu.pl.axvline(x=0., color = 'k', ls='--') #mpu.legend(loc='upper center', display_mode = 'less_space', draw_frame = True) mpu.legend(loc='center left', display_mode = 'less_space', draw_frame = True) mpu.figure() mn = np.mean(all_vecs, 1).A1 mn = mn/np.linalg.norm(mn) mpu.plot_yx(mn, color = '#FF3300', label = 'mean (normalized)', scatter_size=7) c_list = ['#00CCFF', '#643DFF'] for i in range(2): mpu.plot_yx(U[:,i].flatten(), color = c_list[i], label = 'Eigenvector %d'%(i+1), scatter_size=7) mpu.pl.axhline(y=0., color = 'k', ls='--') mpu.legend(display_mode = 'less_space', draw_frame=False) mpu.show() ## # makes a scatter plot with the radius along the x axis and the max # force along the y-axis. different color for each mechanism # @param dir_name - directory containing the pkls. def max_force_hist(dir_name_list, open=True, type=''): if open: trial = 'Opening' else: trial = 'Closing' fls, freezer_list, fridge_list, springloaded_list = [],[],[],[] broiler_list = [] num_mech = 0 lens = [] max_angle = math.radians(15.) max_dist = 0.1 for d in dir_name_list: nm = get_mech_name(d) ep = extract_pkls(d, open, ignore_moment_list=True) if ep == None: continue ftan_l_l = ep['ftan_l_l'] frad_l_l = ep['frad_l_l'] mechx_l_l = ep['mechx_l_l'] typ = ep['typ'] rad = ep['rad'] fl, rl = [], [] for n in range(len(ftan_l_l)): ftan_l = ftan_l_l[n] mechx_a = np.array(mechx_l_l[n]) if type == 'prismatic': indices = np.where(mechx_a < max_dist)[0] else: indices = np.where(mechx_a < max_angle)[0] if len(indices) > 0: ftan_l = np.array(ftan_l)[indices].tolist() fmax = np.max(np.abs(ftan_l)) fl.append(fmax) rl.append(rad) #fmax_max = np.max(fl) fmax_max = np.min(fl) if type == 'rotary': if 'ree' in nm: freezer_list.append(fmax_max) elif 'naveen_microwave' in nm: # putting microwave in freezers freezer_list.append(fmax_max) if fmax_max < 5.: print 'nm:', nm elif 'ge' in nm: fridge_list.append(fmax_max) elif fmax > 60.: springloaded_list.append(fmax_max) else: if 'ven' in nm: broiler_list.append(fmax_max) if fmax_max > 10.: print 'nm:', nm, 'fmax:', fmax_max fls.append(fmax_max) num_mech += 1 lens.append(len(fl)) if len(fls) > 0: max_force = np.max(fls) bin_width = 2.5 bins = np.arange(0.-bin_width/2., max_force+2*bin_width, bin_width) if type == 'rotary': mpu.set_figure_size(3.,4.) mpu.figure() hist, bin_edges = np.histogram(fls, bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Force (Newtons)', ylabel='\# of mechanisms', color='b', label='Cabinets') max_freq = np.max(hist) hist, bin_edges = np.histogram(freezer_list + fridge_list, bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Force (Newtons)', ylabel='\# of mechanisms', color='y', label='Appliances') hist, bin_edges = np.histogram(springloaded_list, bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Force (Newtons)', ylabel='\# of mechanisms', color='g', label='Spring Loaded Doors') mpu.pl.xticks(np.arange(0.,max_force+2*bin_width, 10.)) mpu.legend(display_mode='less_space', handlelength=1.) pb.xlim(-bin_width, max_force+bin_width) pb.ylim(0, max_freq+0.5) else: mpu.set_figure_size(3.,4.) mpu.figure() hist, bin_edges = np.histogram(fls, bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Force (Newtons)', ylabel='\# of mechanisms', color='b', label='Drawers') max_freq = np.max(hist) hist, bin_edges = np.histogram(broiler_list, bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Force (Newtons)', ylabel='\# of mechanisms', color='y', label='Broilers') mpu.figure() bin_width = 2.5 bins = np.arange(0.-bin_width/2., max_force+2*bin_width, bin_width) hist, bin_edges = np.histogram(fls, bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Force (Newtons)', ylabel='\# of mechanisms', color='b', label='All') mpu.legend() pb.xlim(-bin_width, max_force+bin_width) pb.ylim(0, np.max(hist)+1) mpu.pl.xticks(np.arange(0.,max_force+2*bin_width, 5.)) mpu.pl.yticks(np.arange(0.,np.max(hist)+0.5, 1.)) else: print "OH NO! FLS <= 0" def dimen_reduction_mechanisms(dir_list, dimen = 2): mech_vec_list = [] legend_list = [] dir_list = filter_dir_list(dir_list) dir_list_new = [] for d in dir_list: v = make_vector_mechanism(d) if v == None: continue print 'v.shape:', v.shape mech_vec_list.append(v) legend_list.append(get_mech_name(d)) dir_list_new.append(d) all_vecs = np.column_stack(mech_vec_list) #U, s, _ = np.linalg.svd(np.cov(all_vecs)) normalized_all_vecs = all_vecs print '>>>>>>>>>>>> all_vecs.shape <<<<<<<<<<<<<<<', all_vecs.shape #normalized_all_vecs = (all_vecs - all_vecs.mean(1)) #Rule this out as it places equal value on the entire trajectory but we want #to focus modeling efforts on the beginning of the trajectory #normalized_all_vecs = normalized_all_vecs / np.std(normalized_all_vecs, 1) #Removes the effect of force scaling... not sure if we want this #normalized_all_vecs = normalized_all_vecs / ut.norm(normalized_all_vecs) U, s, _ = np.linalg.svd(np.cov((normalized_all_vecs))) perc_account = np.cumsum(s) / np.sum(s) mpu.plot_yx([0]+list(perc_account)) mpu.set_figure_size(5.,5.) mpu.figure() mn = np.mean(all_vecs, 1).A1 mpu.plot_yx(mn/np.linalg.norm(mn), color = '#FF3300', label = 'mean (normalized)', scatter_size=5) c_list = ['#%02x%02x%02x'%(r,g,b) for (r,g,b) in [(152, 32, 176), (23,94,16)]] c_list = ['#00CCFF', '#643DFF'] for i in range(2): mpu.plot_yx(U[:,i].flatten(), color = c_list[i], label = 'Eigenvector %d'%(i+1), scatter_size=5) mpu.pl.axhline(y=0., color = 'k') mpu.legend(display_mode='less_space') if dimen == 2: fig = mpu.figure() proj_mat = U[:, 0:2] for i, v in enumerate(mech_vec_list[:]): p = proj_mat.T * (v - np.matrix(mn).T) color = mpu.random_color() label = legend_list[i] mpu.plot_yx(p[1,:].A1, p[0,:].A1, color = color, linewidth = 0, label = label, xlabel='\huge{First Principle Component}', ylabel='\huge{Second Principle Component}', axis = 'equal', picker=0.5) mpu.pl.axhline(y=0., color = 'k', ls='--') mpu.pl.axvline(x=0., color = 'k', ls='--') mpu.legend() ppg = pca_plot_gui(legend_list, mech_vec_list, U, dir_list_new, np.matrix(mn).T) fig.canvas.mpl_connect('button_press_event', ppg.pick_cb) mpu.show() def filter_dir_list(dir_list, typ = 'rotary', name = None): filt_list = [] for d in dir_list: nm = get_mech_name(d) if name != None: if name not in nm: continue # trial = 'open' # l = glob.glob(d+'/*'+trial+'*mechanism_trajectories*.pkl') # di = ut.load_pickle(l[0]) # m_typ = di['typ'] # if m_typ != typ: # continue # filt_list.append(d) di = extract_pkls(d, quiet=True) if di == None: continue if di.has_key('typ'): m_typ = di['typ'] if m_typ != typ: continue filt_list.append(d) else: continue return filt_list if __name__ == '__main__': import optparse p = optparse.OptionParser() p.add_option('-d', '--dir', action='store', default='', type='string', dest='dir', help='directory with logged data') p.add_option('--check_data', action='store_true', dest='check_data', help='count the number of trajectories for each mechanism') p.add_option('--rearrange', action='store_true', dest='rearrange', help='rearrange aggregated pkls into separate folders for each mechanism') p.add_option('--clean', action='store_true', dest='clean', help='remove pkls with corrupted data') p.add_option('--max_force_hist', action='store_true', dest='max_force_hist', help='histogram of max forces') p.add_option('--max_force_radius_scatter', action='store_true', dest='max_force_radius_scatter', help='scatter plot of max force vs radius') p.add_option('--opening_distances_drawers', action='store_true', dest='opening_distances_drawers', help='opening distances for drawers') p.add_option('--plot_handle_height', action='store_true', dest='plot_handle_height', help='handle height above the ground') p.add_option('--plot_radius', action='store_true', dest='plot_radius', help='histogram of radii of the mechanisms') p.add_option('--correlate', action='store_true', dest='correlate', help='correlation across different trials') p.add_option('--consistent_across_people', action='store_true', dest='consistent_across_people', help='plot mean and std for tangential and total magnitude of the force') p.add_option('--dimen_reduc', action='store_true', dest='dimen_reduc', help='try dimen reduction') p.add_option('--independence', action='store_true', dest='independence', help='test for conditional independence') p.add_option('--mech_models', action='store_true', dest='mech_models', help='fit mechanical models to data') p.add_option('--different_classes_rotary', action='store_true', dest='different_classes_rotary', help='scatter plot showing the differnt mechanism classes') opt, args = p.parse_args() dir_list = commands.getoutput('ls -d %s/*/'%(opt.dir)).splitlines() # drawers = 0 # cabinets = 0 # for d in dir_list: # nm = get_mech_name(d) # di = extract_pkls(d) # if di == None: # print 'di was None for:', nm # continue # if di['rad'] == -1: # drawers += 1 # else: # cabinets += 1 # # print 'drawers:', drawers # print 'cabinets:', cabinets # import sys; sys.exit() if opt.clean: clean_data_forces(opt.dir) elif opt.rearrange: # listing all the different mechanisms pkl_list = commands.getoutput('ls %s/*.pkl'%(opt.dir)).splitlines() mech_name_list = [] for p in pkl_list: nm = '_'.join(p.split('/')[-1].split('_')[:-5]) print 'p:', p print 'nm:', nm mech_name_list.append(nm) mech_name_list = list(set(mech_name_list)) print 'List of unique mechanisms:', mech_name_list for mech_name in mech_name_list: nm = '%s/%s'%(opt.dir, mech_name) os.system('mkdir %s'%nm) os.system('mv %s/*%s*.pkl %s'%(opt.dir, mech_name, nm)) elif opt.check_data: mech_list = [] for i, m in enumerate(dir_list): t = m.split('/') mech = t[-1] if mech == '': mech = t[-2] mech_list.append(mech) print '%d. %s'%(i, mech) for i,d in enumerate(dir_list): mech_nm = mech_list[i] print '------------------------------------' print 'Mechanism name:', mech_nm for trial in ['open', 'close']: l = glob.glob(d+'/*'+trial+'*mechanism_trajectories*.pkl') l.sort() print 'Number of %s trials: %d'%(trial, len(l)) elif opt.max_force_radius_scatter: max_force_radius_scatter(dir_list[0:], open=True) max_force_radius_scatter(dir_list[0:], open=False) mpu.show() elif opt.max_force_hist: print 'Found %d mechanisms' % len(dir_list[0:]) max_force_hist(filter_dir_list(dir_list[0:], typ='rotary'), open=True, type='rotary') max_force_hist(filter_dir_list(dir_list[0:], typ='prismatic'), open=True, type='prismatic') mpu.show() elif opt.opening_distances_drawers: plot_opening_distances_drawers(dir_list[0:]) mpu.show() elif opt.plot_radius: l1 = filter_dir_list(dir_list, name='ree') l2 = filter_dir_list(dir_list, name='ge') print 'LEN:', len(filter_dir_list(dir_list, name=None)) bar1 = mechanism_radius_histogram(filter_dir_list(dir_list, name=None)) bar2 = mechanism_radius_histogram(l1+l2, color='y') labels = ['Other', 'Freezers and Refrigerators'] mpu.pl.legend([bar1[0],bar2[0]], labels, loc='best') mpu.show() elif opt.plot_handle_height: #plot_title = 'Opening force at different heights' #plot_handle_height(dir_list[:], plot_title) #plot_handle_height_no_fridge_no_freezer() # plot_handle_height_no_office() #handle_height_histogram(dir_list, plot_title='Homes excluding \n refrigerators and freezers') #handle_height_histogram(filter_dir_list(dir_list, name='ge'), # plot_title = 'Refrigerators') handle_height_histogram(filter_dir_list(dir_list, name='ree'), plot_title = 'Freezers') mpu.show() elif opt.correlate: cl = [] lab_list = [] ch_list = input_mechanism_list(dir_list) for i, d in enumerate(ch_list): nm = get_mech_name(d) di = extract_pkls(d, open) ftan_l_l = di['ftan_l_l'] frad_l_l = di['frad_l_l'] mechx_l_l = di['mechx_l_l'] time_l_l = di['time_l_l'] typ = di['typ'] rad = di['rad'] #ftan_l_l, frad_l_l, mechx_l_l, typ, rad = extract_pkls(d, open) for j in range(len(ftan_l_l)): if typ == 'rotary': traj_vel = compute_trajectory_velocity(mechx_l_l[j],time_l_l[j],1) if traj_vel > math.radians(30): continue t, f = bin(mechx_l_l[j], ftan_l_l[j], math.radians(1.), max, ignore_empty=True, max_pose=math.radians(60)) if typ == 'prismatic': t, f = bin(mechx_l_l[j], ftan_l_l[j], 0.01, max, ignore_empty=True, max_pose=0.3) cl.append(f) lab_list.append(nm) correlate_trials(cl[0], cl[:], lab_list) mpu.show() elif opt.consistent_across_people: ch_list = input_mechanism_list(dir_list) for dir in ch_list: compare_tangential_total_magnitude(dir) mpu.show() elif opt.different_classes_rotary: different_classes_rotary(dir_list) elif opt.independence: test_independence_mechanism(dir_list) elif opt.dimen_reduc: #ch_list = input_mechanism_list(dir_list) #dimen_reduction_mechanisms(ch_list) #dimen_reduction_mechanisms(filter_dir_list(dir_list, name='HSI_Suite_210_brown_cabinet_right')) dimen_reduction_mechanisms(dir_list) elif opt.mech_models: dir = filter_dir_list(dir_list, name='patient_room_door')[0] d = extract_pkls(dir, True) trial_num_list = d['trial_num_l'] states_l = [] xarr_l = [] varr_l = [] aarr_l = [] forces_l = [] for i, n in enumerate(trial_num_list): print 'i:', i # if i != 5: # continue #trial_idx = trial_num_list.index() mechx_l = d['mechx_l_l'][i] time_list = d['time_l_l'][i] forces = np.matrix(d['ftan_l_l'][i]) radius = d['rad'] window_len = 15 mechx_l, vel, acc, time_list = mfc.kinematic_params(mechx_l, time_list, window_len) xarr = np.array(mechx_l) idxs = np.where(np.logical_and(xarr > math.radians(1), xarr < math.radians(15))) # smooth x forces = forces[0, window_len-1:-window_len+1] #compute vel forces = forces[0, 1:-1] # smooth vel forces = forces[0, window_len-1:-window_len+1] # compute acc forces = forces[0, 1:-1] # smooth acc forces = forces[0, window_len-1:-window_len+1] forces = forces[0, idxs[0]] xarr = np.matrix(xarr[idxs]) varr = np.matrix(np.array(vel)[idxs]) aarr = np.matrix(np.array(acc)[idxs]) xarr_l.append(xarr) varr_l.append(varr) aarr_l.append(aarr) forces_l.append(forces) #import pdb #pdb.set_trace() #door_smd = mfd.DoorSpringMassDamper(radius, k=0., mass=0, damp=0., const=0, noise_cov=0.) door_smd = mfd.DoorMassDamper(radius) ones = np.ones(xarr.shape[1]) states_l.append(np.matrix(np.row_stack((xarr, varr, aarr, ones)))) states = np.column_stack(states_l) forces = np.column_stack(forces_l) acc_arr = states[2,:].A1 farr = forces.A1 # import pdb; pdb.set_trace() print 'farr:', farr print 'acc_arr:', acc_arr mass = np.mean(np.divide(farr, acc_arr)) print 'mass:', mass #door_smd = door_smd.fit(states, forces.T) door_smd = door_smd.fit(states[[1,2],:], forces.T) print 'fitted model', door_smd for i in range(len(xarr_l)): forces = forces_l[i] xarr = xarr_l[i] varr = varr_l[i] aarr = aarr_l[i] predicted_forces = door_smd.predict(states_l[i][[1,2],:]) predicted_forces_mass = mass * states_l[i][2,:].A1 mpu.figure() mpu.plot_yx(forces.A1, np.degrees(xarr.A1), label='actual', color='g') mpu.plot_yx(predicted_forces.A1, np.degrees(xarr.A1), label='predicted SMD', color='b') mpu.plot_yx(predicted_forces_mass, np.degrees(xarr.A1), label='predicted (mass only)', color='r') mpu.plot_yx(np.degrees(varr.A1), np.degrees(xarr.A1), label='vel', color=mpu.random_color()) mpu.plot_yx(np.degrees(aarr.A1), np.degrees(xarr.A1), label='acel', color=mpu.random_color()) mpu.legend() #mpu.show() #plot_velocity(dir) mpu.show() else: # filt_list = filter_dir_list(dir_list) # for dir in filt_list: # max_force_vs_velocity(dir) # mpu.show() d_list = input_mechanism_list(dir_list) # d_list = filter_dir_list(dir_list, typ='prismatic') # mpu.figure() traj_vel_l = [] for dir in d_list: #make_vector_mechanism(dir) # vel_l = [] # nm = get_mech_name(dir) # print '>>>>>>>>>> nm:', nm # d = extract_pkls(dir, ignore_moment_list=True) # for i, mech_x in enumerate(d['mechx_l_l']): # time_list = d['time_l_l'][i] # #v = compute_average_velocity(mech_x, time_list, max_angle=math.radians(30), type='rotary') # v = compute_average_velocity(mech_x, time_list, max_angle=0.2, type='prismatic') # print 'v:', v # if v == -1: # continue # vel_l.append(v) # sorted_vel_l = sorted(vel_l) # if len(sorted_vel_l) > 6: # vel_l = sorted_vel_l[0:6] # else: # vel_l = sorted_vel_l # traj_vel_l += vel_l # # #print 'mean angular velocity:', math.degrees(np.mean(traj_vel_l)) # #print 'std angular velocity:', math.degrees(np.std(traj_vel_l)) # # print 'mean angular velocity:', np.mean(traj_vel_l) # print 'std angular velocity:', np.std(traj_vel_l) #plot_tangential_force(dir, all_trials = True, open = True, # filter_speed = math.radians(30)) plot_tangential_force(dir, all_trials = True, open = True) # nm = get_mech_name(dir) # mpu.savefig(nm+'_grayscale.png') # print 'Aloha ', nm #radial_tangential_ratio(dir) mpu.show()
[ [ 1, 0, 0.0037, 0.0006, 0, 0.66, 0, 760, 0, 1, 0, 0, 760, 0, 0 ], [ 1, 0, 0.0043, 0.0006, 0, 0.66, 0.0222, 688, 0, 1, 0, 0, 688, 0, 0 ], [ 1, 0, 0.0049, 0.0006, 0, ...
[ "import commands", "import os", "import os.path as pt", "import glob", "import math, numpy as np", "import scipy.signal as ss", "import scipy.cluster as clus", "import roslib; roslib.load_manifest('modeling_forces')", "import roslib; roslib.load_manifest('modeling_forces')", "import modeling_force...
import roslib; roslib.load_manifest('modeling_forces') import rospy import hrl_lib.util as ut import hrl_lib.transforms as tr import matplotlib_util.util as mpu import hrl_tilting_hokuyo.display_3d_mayavi as d3m import modeling_forces.smooth as mfs import kinematics_estimation as ke import glob import math, numpy as np import sys ## # plot to ensure that the time stamps in the different logs are # reasonable. # TODO - check for the rates too. def check_time_sync(ft_time_list, mechanism_time_list, hand_time_list): mpu.plot_yx(np.zeros(len(ft_time_list))+1, ft_time_list, color = mpu.random_color(), label='ft\_time\_list', axis=None, linewidth=0.5, scatter_size=10) mpu.plot_yx(np.zeros(len(mechanism_time_list))+2, mechanism_time_list, color = mpu.random_color(), label='mechanism\_time\_list', axis=None, linewidth=0.5, scatter_size=10) mpu.plot_yx(np.zeros(len(hand_time_list))+3, hand_time_list, color = mpu.random_color(), label='hand\_time\_list', axis=None, linewidth=0.5, scatter_size=10) mpu.legend() # mpu.show() ## # # @return single dict with ft_list, mech_pose_lists, hand_pose_lists # and ONE time_list def synchronize(ft_dict, mechanism_dict, hand_dict): ft_time_arr = np.array(ft_dict['time_list']) mech_time_arr = np.array(mechanism_dict['time_list']) hand_time_arr = np.array(hand_dict['time_list']) print 'ft_time_arr.shape:', ft_time_arr.shape print 'mech_time_arr.shape:', mech_time_arr.shape print 'hand_time_arr.shape:', hand_time_arr.shape start_time = max(ft_time_arr[0], mech_time_arr[0], hand_time_arr[0]) end_time = min(ft_time_arr[-1], mech_time_arr[-1], hand_time_arr[-1]) t1_arr = mech_time_arr[np.where(np.logical_and(mech_time_arr >= start_time, mech_time_arr <= end_time))] t2_arr = hand_time_arr[np.where(np.logical_and(hand_time_arr >= start_time, hand_time_arr <= end_time))] #time_arr = np.arange(start_time, end_time, 0.03) # 30ms n_times = min(len(t1_arr), len(t2_arr)) time_arr_list = [] i, j = 0, 0 while True: if t1_arr[i] == t2_arr[j]: time_arr_list.append(t1_arr[i]) i += 1 j += 1 elif t1_arr[i] > t2_arr[j]: j += 1 else: i += 1 if j == n_times or i == n_times: break time_arr = np.array(time_arr_list) tstep_size = .03333333333 uniform_time = np.cumsum(np.round((time_arr[1:] - time_arr[:-1]) / tstep_size) * tstep_size) uniform_time = np.concatenate((np.array([0]), uniform_time)) uniform_time = uniform_time + time_arr_list[0] time_arr = uniform_time # adding a 50ms bias. see modeling_forces/image_ft_sync_test ft_time_arr = ft_time_arr + 0.05 raw_ft_arr = np.array(ft_dict['ft_list']).T window_len = 3 sm_ft_l = [] for i in range(raw_ft_arr.shape[0]): s = mfs.smooth(raw_ft_arr[i,:], window_len,'blackman') sm_ft_l.append(s.tolist()) # smooth truncates the array if window_len != 1: ft_time_arr = ft_time_arr[window_len-1:-window_len+1] raw_ft_arr = (np.array(sm_ft_l).T).tolist() raw_mech_pos_arr = mechanism_dict['pos_list'] raw_mech_rot_arr = mechanism_dict['rot_list'] raw_hand_pos_arr = hand_dict['pos_list'] raw_hand_rot_arr = hand_dict['rot_list'] raw_arr_list = [raw_ft_arr, raw_mech_pos_arr, raw_mech_rot_arr, raw_hand_pos_arr, raw_hand_rot_arr] time_arr_list = [ft_time_arr, mech_time_arr, mech_time_arr, hand_time_arr, hand_time_arr] n_arr = len(raw_arr_list) ft_list = [] mech_pos_list, mech_rot_list = [], [] hand_pos_list, hand_rot_list = [], [] acc_list = [ft_list, mech_pos_list, mech_rot_list, hand_pos_list, hand_rot_list] key_list = ['ft_list', 'mech_pos_list', 'mech_rot_list', 'hand_pos_list', 'hand_rot_list'] for i in range(time_arr.shape[0]): t = time_arr[i] for j in range(n_arr): # nearest neighbor interpolation min_idx = np.argmin(np.abs(time_arr_list[j] - t)) acc_list[j].append(raw_arr_list[j][min_idx]) d = {} d['time_list'] = time_arr.tolist() for i in range(n_arr): d[key_list[i]] = acc_list[i] return d #--------------- functions that operate on combined pkl ---------------------- ## # transform forces to camera coord frame. # @param hand_rot_matrix - rotation matrix for camera to hand checker. # @param hand_pos_matrix - position of hand checkerboard in camera coord frame. # @param mech_pos_matrix - position of mechanism checkerboard in camera coord frame. # @param number - checkerboard number (1, 2, 3 or 4) def ft_to_camera(force_tool, hand_rot_matrix, hand_pos_matrix, mech_pos_matrix, number): # hc == hand checkerboard hc_rot_tool = tr.Rx(math.radians(90)) * tr.Ry(math.radians(180.)) * tr.Rz(math.radians(30.)) while number != 1: hc_rot_tool = tr.Ry(math.radians(90.)) * hc_rot_tool number = number-1 force_hc = hc_rot_tool * force_tool p_hc_ft = np.matrix([0.04, 0.01, 0.09]).T # vector from hook checkerboard origin to the base of the FT sensor in hook checker coordinates. # vec from FT sensor to mechanism checker origin in camera coordinates. p_ft_mech = -hand_pos_matrix + mech_pos_matrix - hand_rot_matrix * p_hc_ft force_cam = hand_rot_matrix * force_hc # force at hook base in camera coordinates moment_cam = hand_rot_matrix * moment_hc force_at_mech_origin = force_cam moment_at_mech_origin = moment_cam + np.matrix(np.cross(-p_ft_mech.A1, force_cam.A1)).T return hand_rot_matrix * force_hc ## # transform force to camera coord frame. def ft_to_camera_3(force_tool, moment_tool, hand_rot_matrix, number, return_moment_cam = False): # hc == hand checkerboard hc_rot_tool = tr.Rx(math.radians(90)) * tr.Ry(math.radians(180.)) * tr.Rz(math.radians(30.)) while number != 1: hc_rot_tool = tr.Ry(math.radians(90.)) * hc_rot_tool number = number-1 force_hc = hc_rot_tool * force_tool moment_hc = hc_rot_tool * moment_tool p_ft_hooktip = np.matrix([0.0, -0.08, 0.00]).T # vector from FT sensor to the tip of the hook in hook checker coordinates. # p_ft_hooktip = np.matrix([0.0, -0.08 - 0.034, 0.00]).T # vector from FT sensor to the tip of the hook in hook checker coordinates. p_ft_hooktip = hand_rot_matrix * p_ft_hooktip force_cam = hand_rot_matrix * force_hc # force at hook base in camera coordinates moment_cam = hand_rot_matrix * moment_hc force_at_hook_tip = force_cam moment_at_hook_tip = moment_cam + np.matrix(np.cross(-p_ft_hooktip.A1, force_cam.A1)).T # if np.linalg.norm(moment_at_hook_tip) > 0.7: # import pdb; pdb.set_trace() if return_moment_cam: return force_at_hook_tip, moment_at_hook_tip, moment_cam else: return force_at_hook_tip, moment_at_hook_tip def plot(combined_dict, savefig): plot_trajectories(combined_dict) # tup = ke.init_ros_node() # mech_rot_list = compute_mech_rot_list(combined_dict, tup) # combined_dict['mech_rot_list'] = mech_rot_list # plot_trajectories(combined_dict) cd = combined_dict ft_mat = np.matrix(cd['ft_list']).T force_mat = ft_mat[0:3, :] mpu.plot_yx(ut.norm(force_mat).A1, cd['time_list']) mpu.show() # plot_forces(combined_dict) # if savefig: # d3m.savefig(opt.dir+'/trajectory_check.png', size=(900, 800)) # else: # d3m.show() ## # @param pts - 3xN np matrix def project_points_plane(pts): # import mdp # p = mdp.nodes.PCANode(svd=True) # p.train((pts-np.mean(pts, 1)).T.A) # print 'min mdp:', p.get_projmatrix() # # U, s , _ = np.linalg.svd(np.cov(pts)) # print 'min svd:', U[:,2] eval, evec = np.linalg.eig(np.cov(pts)) min_idx = np.argmin(eval) min_evec = np.matrix(evec[:,min_idx]).T if min_evec[1,0] > 0: min_evec = min_evec * -1 print 'min evec:', min_evec pts_proj = pts - np.multiply((min_evec.T * pts), min_evec) return pts_proj, min_evec ## # use method 1 to compute the mechanism angle from combined dict. # method 1 - angle between the x axis of checkerboard coord frame. # @return list of mechanism angles. def compute_mech_angle_1(cd, axis_direc=None): mech_rot = cd['mech_rot_list'] directions_x = (np.row_stack(mech_rot)[:,0]).T.reshape(len(mech_rot), 3).T if axis_direc != None: directions_x = directions_x - np.multiply(axis_direc.T * directions_x, axis_direc) directions_x = directions_x/ut.norm(directions_x) start_normal = directions_x[:,0] mech_angle = np.arccos(start_normal.T * directions_x).A1.tolist() return mech_angle ## # use method 2 to compute the mechanism angle from combined dict. # method 2 - fit a circle to estimate location of axis of rotation and # radius. Then use that to compute the angle of the mechanism. # @return list of mechanism angles. def compute_mech_angle_2(cd, tup, project_plane=False): pos_mat = np.column_stack(cd['mech_pos_list']) if project_plane: pts_proj, min_evec = project_points_plane(pos_arr) pos_arr = pts_proj kin_dict = ke.fit(pos_mat, tup) center = np.matrix((kin_dict['cx'], kin_dict['cy'], kin_dict['cz'])).T directions_x = pos_mat - center directions_x = directions_x / ut.norm(directions_x) start_normal = directions_x[:,0] #mech_angle = np.arccos(start_normal.T * directions_x).A1.tolist() ct = (start_normal.T * directions_x).A1 st = ut.norm(np.matrix(np.cross(start_normal.A1, directions_x.T.A)).T).A1 mech_angle = np.arctan2(st, ct).tolist() return mech_angle def compute_mech_rot_list(cd, tup, project_plane=False): pos_arr = np.column_stack(cd['mech_pos_list']) rot_list = cd['mech_rot_list'] directions_y = (np.row_stack(rot_list)[:,1]).T.reshape(len(rot_list), 3).T start_y = directions_y[:,0] pts_proj, y_mech = project_points_plane(pos_arr) if np.dot(y_mech.T, start_y.A1) < 0: print 'Negative hai boss' y_mech = -1 * y_mech if project_plane: pos_arr = pts_proj kin_dict = ke.fit(np.matrix(pos_arr), tup) center = np.array((kin_dict['cx'], kin_dict['cy'], kin_dict['cz'])).T print 'pos_arr[:,0]', pos_arr[:,0] print 'center:', center directions_x = (np.row_stack(rot_list)[:,0]).T.reshape(len(rot_list), 3).T start_x = directions_x[:,0] directions_x = np.matrix(pos_arr) - np.matrix(center).T.A directions_x = directions_x / ut.norm(directions_x) if np.dot(directions_x[:,0].A1, start_x.A1) < 0: print 'Negative hai boss' directions_x = -1 * directions_x mech_rot_list = [] for i in range(len(rot_list)): x = -directions_x[:, i] y = np.matrix(y_mech) z = np.matrix(np.cross(x.A1, y.A1)).T rot_mat = np.column_stack((x, y, z)) mech_rot_list.append(rot_mat) # return mech_rot_list return rot_list ## # @param tup - if None then use method 1 else use method 2 to # compute mech angle. # @return 1d array (radial force), 1d array (tangential force), list of mechanism angles, type ('rotary' or 'prismatic') def compute_mechanism_properties(combined_dict, bias_ft = False, tup = None, cd_pkl_name = None): cd = combined_dict force_cam, moment_cam, _ = fts_to_camera(combined_dict) moment_contact_l = None if bias_ft: print 'Biasing magnitude:', np.linalg.norm(force_cam[:,0]) force_cam = force_cam - force_cam[:,0] moment_cam = moment_cam - moment_cam[:,0] if cd['radius'] != -1: if tup == None: mech_angle = compute_mech_angle_1(cd) else: mech_angle = compute_mech_angle_2(cd, tup) # compute new mech_rot_list. used for factoring the forces. mech_rot_list = compute_mech_rot_list(combined_dict, tup) combined_dict['mech_rot_list'] = mech_rot_list hook_tip_l = compute_hook_tip_trajectory(cd) hand_mat = np.column_stack(hook_tip_l) for i,f in enumerate(force_cam.T): fmag = np.linalg.norm(f) if fmag > 1.0: break end_idx = np.argmax(mech_angle) hand_mat_short = hand_mat[:,i:end_idx] kin_dict = ke.fit(hand_mat_short, tup) center_hand = np.matrix((kin_dict['cx'], kin_dict['cy'], kin_dict['cz'])).T radius_hand = kin_dict['r'] center_mech_coord = mech_rot_list[0].T * center_hand start_mech_coord = mech_rot_list[0].T * hand_mat_short[:,0] opens_right = False if start_mech_coord[0,0] < center_mech_coord[0,0]: print 'Opens Right' opens_right = True # radial vectors. radial_mat = hand_mat - center_hand radial_mat = radial_mat / ut.norm(radial_mat) _, nrm_hand = project_points_plane(hand_mat_short) if np.dot(nrm_hand.A1, mech_rot_list[0][:,1].A1) < 0: nrm_hand = -1 * nrm_hand if opens_right == False: nrm_hand = -1 * nrm_hand frad_l = [] ftan_l = [] moment_contact_l = [] for i, f in enumerate(force_cam.T): f = f.T m = (moment_cam[:,i].T * nrm_hand)[0,0] moment_contact_l.append(m) tvec = np.matrix(np.cross(nrm_hand.A1, radial_mat[:,i].A1)).T ftan = (f.T * tvec)[0,0] ftan_l.append(ftan) frad = np.linalg.norm(f - tvec * ftan) #frad = (f_cam.T*radial_mat[:,i])[0,0] frad_l.append(abs(frad)) typ = 'rotary' else: pos_mat = np.column_stack(cd['mech_pos_list']) mech_angle = ut.norm(pos_mat-pos_mat[:,0]).A1.tolist() #print 'mech_angle:', mech_angle typ = 'prismatic' moment_axis_list = None frad_l = [] ftan_l = [] moment_contact_l = [] rot_list = cd['mech_rot_list'] directions_z = (np.row_stack(rot_list)[:,2]).T.reshape(len(rot_list), 3).T for i, f in enumerate(force_cam.T): f = f.T tvec = np.matrix(directions_z[:,i]) ftan = (f.T * tvec)[0,0] ftan_l.append(ftan) frad = np.linalg.norm(f - tvec * ftan) #frad = (f_cam.T*radial_mat[:,i])[0,0] frad_l.append(abs(frad)) radius_hand = 10. ut.save_pickle(combined_dict, cd_pkl_name) return np.array(frad_l), np.array(ftan_l), mech_angle, typ, \ np.array(ftan_l)*radius_hand, np.array(moment_contact_l) def plot_radial_tangential(mech_dict, savefig, fig_name=''): radial_mech = mech_dict['force_rad_list'] tangential_mech = mech_dict['force_tan_list'] typ = mech_dict['mech_type'] mech_x = mech_dict['mechanism_x'] if typ == 'rotary': mech_x_degrees = np.degrees(mech_x) xlabel = 'angle (degrees)' else: mech_x_degrees = mech_x xlabel = 'distance (meters)' mpu.pl.clf() mpu.plot_yx(radial_mech, mech_x_degrees, axis=None, label='radial force', xlabel=xlabel, ylabel='N', color='r') mpu.plot_yx(tangential_mech, mech_x_degrees, axis=None, label='tangential force', xlabel=xlabel, ylabel='N', color='g') mpu.legend() if typ == 'rotary': mpu.figure() rad = mech_dict['radius'] torques_1 = rad * np.array(tangential_mech) torques_3 = np.array(mech_dict['moment_tip_list']) + torques_1 mpu.plot_yx(torques_1, mech_x_degrees, axis=None, label='torque from tangential', xlabel=xlabel, ylabel='moment', color='r') mpu.plot_yx(torques_3, mech_x_degrees, axis=None, label='total torque', xlabel=xlabel, ylabel='moment', color='y') mpu.legend() if savefig: mpu.savefig(opt.dir+'/%s_force_components.png'%fig_name) else: mpu.show() ## # returns force and moment at the tip of the hook in camera # coordinates. def fts_to_camera(combined_dict): cd = combined_dict number = cd['hook_checker_number'] hand_rot = cd['hand_rot_list'] hand_pos = cd['hand_pos_list'] ft_mat = np.matrix(cd['ft_list']).T # 6xN np matrix force_mat = ft_mat[0:3, :] moment_mat = ft_mat[3:6, :] n_forces = force_mat.shape[1] force_cam_list = [] moment_cam_list = [] moment_base_list = [] for i in range(n_forces): f,m,m_base = ft_to_camera_3(force_mat[:,i], moment_mat[:,i], hand_rot[i], number, return_moment_cam = True) force_cam_list.append(f) moment_cam_list.append(m) moment_base_list.append(m_base) force_cam = np.column_stack(force_cam_list) moment_cam = np.column_stack(moment_cam_list) moment_base = np.column_stack(moment_base_list) return force_cam, moment_cam, moment_base def plot_forces(combined_dict): cd = combined_dict hand_mat = np.column_stack(cd['hand_pos_list']) hand_rot = cd['hand_rot_list'] mech_mat = np.column_stack(cd['mech_pos_list']) mech_rot = cd['mech_rot_list'] directions_x = (np.row_stack(mech_rot)[:,0]).T.reshape(len(mech_rot), 3).T force_cam, moment_cam, _ = fts_to_camera(combined_dict) d3m.plot_points(hand_mat, color = (1.,0.,0.), mode='sphere') d3m.plot_points(mech_mat, color = (0.,0.,1.), mode='sphere') d3m.plot_normals(mech_mat, directions_x, color=(1.,0,0.)) # d3m.plot_normals(mech_mat, force_mat, color=(0.,1,0.)) d3m.plot_normals(mech_mat, force_cam, color=(0.,0,1.)) def plot_trajectories(combined_dict): cd = combined_dict hand_mat = np.column_stack(cd['hand_pos_list']) hand_rot = cd['hand_rot_list'] directions_x = (np.row_stack(hand_rot)[:,0]).T.reshape(len(hand_rot), 3).T directions_y = (np.row_stack(hand_rot)[:,1]).T.reshape(len(hand_rot), 3).T directions_z = (np.row_stack(hand_rot)[:,2]).T.reshape(len(hand_rot), 3).T #d3m.white_bg() d3m.plot_points(hand_mat, color = (1.,0.,0.), mode='sphere', scale_factor = 0.005) d3m.plot_normals(hand_mat, directions_x, color=(1.,0,0.), scale_factor = 0.02) d3m.plot_normals(hand_mat, directions_y, color=(0.,1,0.), scale_factor = 0.02) d3m.plot_normals(hand_mat, directions_z, color=(0.,0,1.), scale_factor = 0.02) mech_mat = np.column_stack(cd['mech_pos_list']) mech_rot = cd['mech_rot_list'] directions_x = (np.row_stack(mech_rot)[:,0]).T.reshape(len(mech_rot), 3).T directions_y = (np.row_stack(mech_rot)[:,1]).T.reshape(len(hand_rot), 3).T directions_z = (np.row_stack(mech_rot)[:,2]).T.reshape(len(mech_rot), 3).T d3m.plot_points(mech_mat[:,0:1], color = (0.,0.,0.), mode='sphere', scale_factor = 0.01) d3m.plot_points(mech_mat, color = (0.,0.,1.), mode='sphere', scale_factor = 0.005) d3m.plot_normals(mech_mat, directions_x, color=(1.,0,0.), scale_factor = 0.02) d3m.plot_normals(mech_mat, directions_y, color=(0.,1,0.), scale_factor = 0.02) d3m.plot_normals(mech_mat, directions_z, color=(0.,0,1.), scale_factor = 0.02) m = np.mean(mech_mat, 1) d3m.mlab.view(azimuth=-120, elevation=60, distance=1.60, focalpoint=(m[0,0], m[1,0], m[2,0])) ## # @return list of hook tip coodinates in camera coordinate frame. def compute_hook_tip_trajectory(combined_dict): cd = combined_dict hand_mat = np.column_stack(cd['hand_pos_list']) hand_rot_l = cd['hand_rot_list'] directions_x = (np.row_stack(hand_rot_l)[:,0]).T.reshape(len(hand_rot_l), 3).T directions_y = (np.row_stack(hand_rot_l)[:,1]).T.reshape(len(hand_rot_l), 3).T directions_z = (np.row_stack(hand_rot_l)[:,2]).T.reshape(len(hand_rot_l), 3).T hand_pos_list = cd['hand_pos_list'] hook_tip_l = [] for i,p in enumerate(hand_pos_list): # p - hook checker origin in camera coordinates. hc_P_hc_hooktip = np.matrix([0.035, -0.0864, 0.09]).T # vector from hook checkerboard origin to the tip of the hook in hook checker coordinates. cam_P_hc_hooktip = hand_rot_l[i] * hc_P_hc_hooktip hook_tip_l.append(p + cam_P_hc_hooktip) return hook_tip_l ## # take the open + close trajectory and split it into two separate # trajectories and save them as pkls. def split_open_close(rad, tan, ang, typ, mech_radius, time_list, moment_axis, moment_tip): ang = np.array(ang) incr = ang[1:] - ang[:-1] n_pts = ang.shape[0] - 2 #ignoring first and last readings. rad_l, tan_l, ang_l = [], [], [] for i in range(n_pts): if typ == 'rotary': sgn = incr[i] * incr[i+1] mag = abs(incr[i] - incr[i+1]) if sgn < 0 and mag > math.radians(10): continue rad_l.append(rad[i+1]) tan_l.append(tan[i+1]) ang_l.append(ang[i+1]) else: # no cleanup for prismatic joints, for now rad_l.append(rad[i+1]) tan_l.append(tan[i+1]) ang_l.append(ang[i+1]) rad, tan, ang = rad_l, tan_l, ang_l max_idx = np.argmax(ang) d_open = {'force_rad_list': rad[:max_idx+1], 'force_tan_list': tan[:max_idx+1], 'mechanism_x': ang[:max_idx+1], 'mech_type': typ, 'radius': mech_radius, 'time_list': time_list[:max_idx+1]} if moment_tip != None: d_open['moment_tip_list'] = moment_tip[:max_idx+1] d_open['moment_list'] = moment_axis[:max_idx+1] ut.save_pickle(d_open, opt.dir + '/open_mechanism_trajectories_handhook.pkl') d_close = {'force_rad_list': rad[max_idx+1:], 'force_tan_list': tan[max_idx+1:], 'mechanism_x': ang[max_idx+1:], 'mech_type': typ, 'radius': mech_radius, 'time_list': time_list[max_idx+1:]} if moment_tip != None: d_open['moment_tip_list'] = moment_tip[max_idx+1:] d_open['moment_list'] = moment_axis[max_idx+1:] ut.save_pickle(d_close, opt.dir + '/close_mechanism_trajectories_handhook.pkl') def plot_hooktip_trajectory_and_force(cd): hook_tip_l = compute_hook_tip_trajectory(cd) # plot trajectory in 3D. d3m.white_bg() d3m.plot_points(np.column_stack(hook_tip_l), color = (1.,0.,0.), mode='sphere', scale_factor = 0.005) # d3m.plot_points(mech_proj[:,0:1], color = (0.,0.,0.), mode='sphere', # scale_factor = 0.01) # d3m.plot_points(mech_proj, color = (0.,0.,1.), mode='sphere', # scale_factor = 0.005) # d3m.plot(np.column_stack((mech_proj[:,-1],center_mech, mech_proj[:,0])), # color = (0.,0.,1.)) # d3m.plot(np.column_stack((hand_proj[:,-1],center_hand, hand_proj[:,0])), # color = (1.,0.,0.)) d3m.show() ## # sanity check - fitting circle to mechanism and hook tip # trajectories, computing the angle between the initial radial # direction of the mechanism and the radial directions for the hook # tip. This angle starts out at a slightly positive angle. I'm # assuming that this corresponds to the fact that the handle sticks # out from the cabinet door. What makes me nervous is that I am still # fitting two different circles to the mechanism and hook # trajectories. def compare_tip_mechanism_trajectories(mech_mat, hand_mat): # hand_proj, nrm_hand = project_points_plane(hand_mat) # mech_proj, nrm_mech = project_points_plane(mech_mat) hand_proj = hand_mat mech_proj = mech_mat kin_dict = ke.fit(hand_proj, tup) print 'kin_dict from hook tip:', kin_dict print 'measured radius:', cd['radius'] center_hand = np.matrix((kin_dict['cx'], kin_dict['cy'], kin_dict['cz'])).T kin_dict = ke.fit(mech_proj, tup) print 'kin_dict from mechanism:', kin_dict center_mech = np.matrix((kin_dict['cx'], kin_dict['cy'], kin_dict['cz'])).T # working with the projected coordinates. directions_hand = hand_proj - center_hand directions_hand = directions_hand / ut.norm(directions_hand) directions_mech = mech_proj - center_mech directions_mech = directions_mech / ut.norm(directions_mech) start_normal = directions_mech[:,0] print 'directions_mech[:,0]', directions_mech[:,0].A1 print 'directions_hand[:,0]', directions_hand[:,0].A1 ct = (start_normal.T * directions_hand).A1 st = ut.norm(np.matrix(np.cross(start_normal.A1, directions_hand.T.A)).T).A1 mech_angle = np.arctan2(st, ct).tolist() #mech_angle = np.arccos(start_normal.T * directions_hand).A1.tolist() mpu.plot_yx(np.degrees(mech_angle)) mpu.show() # plot trajectory in 3D. d3m.white_bg() d3m.plot_points(hand_proj, color = (1.,0.,0.), mode='sphere', scale_factor = 0.005) d3m.plot_points(mech_proj[:,0:1], color = (0.,0.,0.), mode='sphere', scale_factor = 0.01) d3m.plot_points(mech_proj, color = (0.,0.,1.), mode='sphere', scale_factor = 0.005) d3m.plot(np.column_stack((mech_proj[:,-1],center_mech, mech_proj[:,0])), color = (0.,0.,1.)) d3m.plot(np.column_stack((hand_proj[:,-1],center_hand, hand_proj[:,0])), color = (1.,0.,0.)) d3m.show() def angle_between_hooktip_mechanism_radial_vectors(mech_mat, hand_mat): kin_dict = ke.fit(hand_mat, tup) print 'kin_dict from hook tip:', kin_dict print 'measured radius:', cd['radius'] center_hand = np.matrix((kin_dict['cx'], kin_dict['cy'], kin_dict['cz'])).T kin_dict = ke.fit(mech_mat, tup) print 'kin_dict from mechanism:', kin_dict center_mech = np.matrix((kin_dict['cx'], kin_dict['cy'], kin_dict['cz'])).T # working with the projected coordinates. directions_hand = hand_mat - center_hand directions_hand = directions_hand / ut.norm(directions_hand) directions_mech = mech_mat - center_mech directions_mech = directions_mech / ut.norm(directions_mech) #import pdb; pdb.set_trace() ang = np.degrees(np.arccos(np.sum(np.multiply(directions_mech, directions_hand), 0))).A1 mpu.plot_yx(ang, label = 'angle between hooktip-radial and mechanism radial') mpu.legend() mpu.show() def split_forces_hooktip_test(hand_mat): kin_dict = ke.fit(hand_mat, tup) center_hand = np.matrix((kin_dict['cx'], kin_dict['cy'], kin_dict['cz'])).T print 'kin_dict:', kin_dict # radial vectors. radial_mat = hand_mat - center_hand radial_mat = radial_mat / ut.norm(radial_mat) # cannot use hook tip to compute mechanism angle because I # don't have a way of knowing when the hook starts opening the # mechanism. (Think hook makes contact with door, moves in # freespace and then makes contact with the handle.) #start_rad = radial_mat[:,0] #ct = (start_rad.T * radial_mat).A1 #st = ut.norm(np.matrix(np.cross(start_rad.A1, radial_mat.T.A)).T).A1 #mech_angle_l = np.arctan2(st, ct).tolist() _, nrm_hand = project_points_plane(hand_mat) print 'nrm_hand:', nrm_hand.A1 f_cam_l = [] m_cam_l = [] m_base_l = [] frad_l = [] ftan_l = [] hook_num = cd['hook_checker_number'] print 'hook_num:', hook_num for i, f in enumerate(force_mat.T): f = f.T m = moment_mat[:,i] f_cam, m_cam, m_base = ft_to_camera_3(f, m, hook_rot_l[i], hook_num, return_moment_cam = True) f_cam_l.append(f_cam) m_cam_l.append(abs((m_cam.T*nrm_hand)[0,0])) m_base_l.append(abs((m_base.T*nrm_hand)[0,0])) #m_base_l.append(np.linalg.norm(f)) tangential_vec = np.matrix(np.cross(radial_mat[:,i].A1, nrm_hand.A1)).T ftan = (f_cam.T * tangential_vec)[0,0] ftan_l.append(ftan) #frad = np.linalg.norm(f_cam - tangential_vec * ftan) frad = (f_cam.T*radial_mat[:,i])[0,0] frad_l.append(abs(frad)) fig1 = mpu.figure() mech_ang_deg = np.degrees(mech_angle_l) mpu.plot_yx(ftan_l, mech_ang_deg, label='Tangential Force (hook tip)', color='b') mpu.plot_yx(frad_l, mech_ang_deg, label='Radial Force (hook tip)', color='y') mech_pkl_name = glob.glob(opt.dir + '/open_mechanism_trajectories_*.pkl')[0] md = ut.load_pickle(mech_pkl_name) radial_mech = md['force_rad_list'] tangential_mech = md['force_tan_list'] mech_x = np.degrees(md['mechanism_x']) mpu.plot_yx(tangential_mech, mech_x, label='Tangential Force (mechanism checker)', color='g') mpu.plot_yx(radial_mech, mech_x, label='Radial Force (mechanism checker)', color='r') mpu.legend() fig2 = mpu.figure() mpu.plot_yx(m_cam_l, mech_ang_deg, label='\huge{$m_{axis}$}') mpu.plot_yx(m_base_l, mech_ang_deg, label='\huge{$m^s$}', color = 'r') mpu.legend() mpu.show() if __name__ == '__main__': import optparse p = optparse.OptionParser() p.add_option('-d', '--dir', action='store', default='', type='string', dest='dir', help='directory with logged data') p.add_option('-t', '--time_check', action='store_true', dest='tc', help='plot to check the consistency of time stamps') p.add_option('-s', '--sync', action='store_true', dest='sync', help='time synchronize poses, forces etc.') p.add_option('--split', action='store_true', dest='split_forces', help='split forces into radial and tangential and save in a pickle') p.add_option('--savefig', action='store_true', dest='savefig', help='save the plot instead of showing it.') p.add_option('-c', '--cd', action='store_true', dest='cd', help='work with the combined dict') p.add_option('-f', '--force', action='store_true', dest='force', help='plot radial and tangential force') p.add_option('--mech_prop_ros', action='store_true', dest='mech_prop_ros', help='plot radial and tangential force') p.add_option('--moment_test', action='store_true', dest='moment_test', help='trying to compute moment about the joint axis.') p.add_option('--hook_tip_test', action='store_true', dest='hook_tip_test', help='plot trajectory of hook tip for debugging etc.') opt, args = p.parse_args() if opt.dir == '': raise RuntimeError('Need a directory to work with (-d or --dir)') if opt.force: mech_pkl_name = glob.glob(opt.dir + '/open_mechanism_trajectories_*.pkl')[0] md = ut.load_pickle(mech_pkl_name) plot_radial_tangential(md, opt.savefig, 'open') # mech_pkl_name = glob.glob(opt.dir + '/close_mechanism_trajectories_handhook.pkl')[0] # md = ut.load_pickle(mech_pkl_name) # plot_radial_tangential(md, opt.savefig, 'close') ft_pkl = glob.glob(opt.dir + '/ft_log*.pkl')[0] poses_pkl = glob.glob(opt.dir + '/poses_dict*.pkl')[0] ft_dict = ut.load_pickle(ft_pkl) poses_dict = ut.load_pickle(poses_pkl) mechanism_dict = poses_dict['mechanism'] hand_dict = poses_dict['hand'] ft_time_list = ft_dict['time_list'] mechanism_time_list = mechanism_dict['time_list'] hand_time_list = hand_dict['time_list'] if opt.tc: check_time_sync(ft_time_list, mechanism_time_list, hand_time_list) if opt.savefig: mpu.savefig(opt.dir+'/time_check.png') else: mpu.show() if opt.sync: print 'Begin synchronize' d = synchronize(ft_dict, mechanism_dict, hand_dict) print 'End synchronize' #ut.save_pickle(d, opt.dir+'/combined_log'+ut.formatted_time()+'.pkl') ut.save_pickle(d, opt.dir+'/combined_log.pkl') print 'Saved pickle' if opt.cd: cd = ut.load_pickle(glob.glob(opt.dir + '/combined_log*.pkl')[0]) plot(cd, opt.savefig) if opt.mech_prop_ros: import mechanism_analyse as ma cd = ut.load_pickle(glob.glob(opt.dir + '/combined_log*.pkl')[0]) tup = ke.init_ros_node() ma2 = compute_mech_angle_2(cd, tup, project_plane=False) ma1 = compute_mech_angle_1(cd) ma3 = compute_mech_angle_2(cd, tup, project_plane=True) # ma4 = compute_mech_angle_1(cd, min_evec) lab1 = 'orientation only' lab2 = 'checker origin position + circle fit' lab3 = 'checker origin position + PCA projection + circle fit' # lab4 = 'PCA projection + orientation' mpu.figure() mpu.plot_yx(np.degrees(ma3), color='r', label=lab3, linewidth = 1, scatter_size = 5) mpu.plot_yx(np.degrees(ma2), color='b', label=lab2, linewidth = 1, scatter_size = 5) mpu.plot_yx(np.degrees(ma1), color='y', label=lab1, linewidth = 1, scatter_size = 5) mpu.legend() vel3 = ma.compute_velocity(ma3, cd['time_list'], 1) vel2 = ma.compute_velocity(ma2, cd['time_list'], 1) vel1 = ma.compute_velocity(ma1, cd['time_list'], 1) mpu.figure() mpu.plot_yx(np.degrees(vel3), np.degrees(ma3), color='r', label=lab3, linewidth = 1, scatter_size = 5) mpu.plot_yx(np.degrees(vel2), np.degrees(ma2), color='b', label=lab2, linewidth = 1, scatter_size = 5) mpu.plot_yx(np.degrees(vel1), np.degrees(ma1), color='y', label=lab1, linewidth = 1, scatter_size = 5) mpu.legend() # acc3 = ma.compute_velocity(vel3, cd['time_list'], 1) # mpu.figure() # mpu.plot_yx(np.degrees(acc3), np.degrees(ma3), color='r', # label=lab3, linewidth = 1, scatter_size = 5) # mpu.legend() mpu.show() if opt.split_forces: tup = ke.init_ros_node() pkl_name = glob.glob(opt.dir + '/combined_log*.pkl')[0] mech_pkl_name = glob.glob(opt.dir + '/mechanism_info*.pkl')[0] md = ut.load_pickle(mech_pkl_name) cd = ut.load_pickle(pkl_name) cd['hook_checker_number'] = md['checkerboard_number'] cd['radius'] = md['radius'] rad, tan, ang, typ, moment_axis, moment_tip = compute_mechanism_properties(cd, bias_ft=True, tup=tup, cd_pkl_name = pkl_name) split_open_close(rad, tan, ang, typ, md['radius'], cd['time_list'], moment_axis, moment_tip) if opt.moment_test: tup = ke.init_ros_node() pkl_name = glob.glob(opt.dir + '/combined_log*.pkl')[0] mech_pkl_name = glob.glob(opt.dir + '/mechanism_info*.pkl')[0] md = ut.load_pickle(mech_pkl_name) cd = ut.load_pickle(pkl_name) cd['hook_checker_number'] = md['checkerboard_number'] cd['radius'] = md['radius'] rad, tan, ang, typ, moment_axis, moment_tip = compute_mechanism_properties(cd, bias_ft=True, tup=tup, cd_pkl_name = pkl_name) ang = np.array(ang) incr = ang[1:] - ang[:-1] n_pts = ang.shape[0] - 2 #ignoring first and last readings. rad_l, tan_l, ang_l = [], [], [] for i in range(n_pts): if typ == 'rotary': sgn = incr[i] * incr[i+1] mag = abs(incr[i] - incr[i+1]) if sgn < 0 and mag > math.radians(10): continue rad_l.append(rad[i+1]) tan_l.append(tan[i+1]) ang_l.append(ang[i+1]) else: # no cleanup for prismatic joints, for now rad_l.append(rad[i+1]) tan_l.append(tan[i+1]) ang_l.append(ang[i+1]) rad, tan, ang = rad_l, tan_l, ang_l max_idx = np.argmax(ang) rad = np.array(rad[:max_idx+1]) tan = np.array(tan[:max_idx+1]) ang = np.array(ang[:max_idx+1]) moment_axis = np.array(moment_axis[:max_idx+1]) moment_tip = np.array(moment_tip[:max_idx+1]) fig1 = mpu.figure() mpu.plot_yx(tan * cd['radius'], np.degrees(ang), label = 'Moment from Tangential Force', color = 'b') mpu.plot_yx(moment_axis, np.degrees(ang), label = 'Computed Moment', color = 'g') mpu.plot_yx(moment_tip, np.degrees(ang), label = 'Computed Moment using tip model', color = 'y') mpu.legend() mpu.show() if opt.hook_tip_test: tup = ke.init_ros_node() pkl_name = glob.glob(opt.dir + '/combined_log*.pkl')[0] mech_pkl_name = glob.glob(opt.dir + '/mechanism_info*.pkl')[0] md = ut.load_pickle(mech_pkl_name) cd = ut.load_pickle(pkl_name) cd['hook_checker_number'] = md['checkerboard_number'] cd['radius'] = md['radius'] hook_tip_l = compute_hook_tip_trajectory(cd) hook_rot_l = cd['hand_rot_list'] mech_mat = np.column_stack(cd['mech_pos_list']) hand_mat = np.column_stack(hook_tip_l) ft_mat = np.matrix(cd['ft_list']).T # 6xN np matrix force_mat = ft_mat[0:3, :] # force_mat = force_mat - force_mat[:,0] moment_mat = ft_mat[3:6, :] # moment_mat = moment_mat - moment_mat[:,0] for i,f in enumerate(force_mat.T): fmag = np.linalg.norm(f) if fmag > 1.0: print 'i:', i break mech_angle_l = compute_mech_angle_2(cd, tup, project_plane=False) end_idx = np.argmax(mech_angle_l) hand_mat = hand_mat[:,i:end_idx] mech_mat = mech_mat[:,i:end_idx] force_mat = force_mat[:,i:end_idx] moment_mat = moment_mat[:,i:end_idx] hook_rot_l = hook_rot_l[i:end_idx] mech_angle_l = mech_angle_l[i:end_idx] compare_tip_mechanism_trajectories(mech_mat[:,i:], hand_mat[:,i:]) #angle_between_hooktip_mechanism_radial_vectors(mech_mat, hand_mat) #plot moment at hook ti and base. #split_forces_hooktip_test(hand_mat)
[ [ 1, 0, 0.002, 0.001, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.002, 0.001, 0, 0.66, 0.0312, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.003, 0.001, 0, 0.66, ...
[ "import roslib; roslib.load_manifest('modeling_forces')", "import roslib; roslib.load_manifest('modeling_forces')", "import rospy", "import hrl_lib.util as ut", "import hrl_lib.transforms as tr", "import matplotlib_util.util as mpu", "import hrl_tilting_hokuyo.display_3d_mayavi as d3m", "import modeli...
# # Subscribe to ObjectDetection message and convert a pickle. # # # import roslib; roslib.load_manifest('modeling_forces') import rospy from checkerboard_detector.msg import ObjectDetection import hrl_lib.util as ut import hrl_lib.transforms as tr from std_msgs.msg import Empty import numpy as np import time def pose_cb(data, d): global t_pose_cb t_pose_cb = time.time() for obj in data.objects: p = obj.pose.position pos = np.matrix([p.x, p.y, p.z]).T q = obj.pose.orientation rot = tr.quaternion_to_matrix([q.x, q.y, q.z, q.w]) t = data.header.stamp.to_sec() d[obj.type]['pos_list'].append(pos) d[obj.type]['rot_list'].append(rot) d[obj.type]['time_list'].append(t) print '>>>>>>>>>>>>>>>>>>>>>>>>>>>>' print 'pos:', pos.T print 'rot:', rot def got_trigger_cb(data, d): d['flag'] = True if __name__ == '__main__': print 'Hello World.' rospy.init_node('checkerboard_poses_to_pickle', anonymous=False) topic_name = '/checkerdetector/ObjectDetection' d = {'mechanism': {}, 'hand': {}} d['mechanism'] = {'pos_list': [], 'rot_list': [], 'time_list': []} d['hand'] = {'pos_list': [], 'rot_list': [], 'time_list': []} rospy.Subscriber(topic_name, ObjectDetection, pose_cb, d) got_trigger_dict = {'flag': False} rospy.Subscriber('/checker_to_poses/trigger', Empty, got_trigger_cb, got_trigger_dict) global t_pose_cb t_pose_cb = time.time() while got_trigger_dict['flag'] == False: time.sleep(0.05) if (time.time() - t_pose_cb) > 60.: raise RuntimeError('haven\'t received a pose_cb in 60 secs') # rospy.spin() print 'Number of poses:', len(d['mechanism']['time_list']) #ut.save_pickle(d, 'poses_dict_'+ut.formatted_time()+'.pkl') ut.save_pickle(d, 'poses_dict.pkl')
[ [ 1, 0, 0.1096, 0.0137, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.1096, 0.0137, 0, 0.66, 0.0909, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.1233, 0.0137, 0, 0....
[ "import roslib; roslib.load_manifest('modeling_forces')", "import roslib; roslib.load_manifest('modeling_forces')", "import rospy", "from checkerboard_detector.msg import ObjectDetection", "import hrl_lib.util as ut", "import hrl_lib.transforms as tr", "from std_msgs.msg import Empty", "import numpy a...
import roslib; roslib.load_manifest('modeling_forces') import rospy from geometry_msgs.msg import Point32 from articulation_msgs.msg import ModelMsg from articulation_msgs.msg import TrackMsg from geometry_msgs.msg import Pose, Point, Quaternion from threading import RLock import numpy as np import time def model_cb(data, kin_dict): print 'model_cb called' if data.name == 'rotational': for p in data.params: if p.name == 'rot_center.x': cx = p.value if p.name == 'rot_center.y': cy = p.value if p.name == 'rot_center.z': cz = p.value if p.name == 'rot_radius': r = p.value kin_dict['typ'] = 'rotational' kin_dict['cx'] = cx kin_dict['cy'] = cy kin_dict['cz'] = cz kin_dict['r'] = r if data.name == 'prismatic': for p in data.params: if p.name == 'prismatic_dir.x': dx = p.value if p.name == 'prismatic_dir.y': dy = p.value if p.name == 'prismatic_dir.z': dz = p.value kin_dict['typ'] = 'prismatic' kin_dict['direc'] = [dx, dy, dz] kin_dict['got_model'] = True ## # fit either a circle or a linear model. # @param pts - 3xN np matrix of points. # @return dictionary with kinematic params. (see model_cb) def fit(pts, tup): kin_dict, track_pub = tup track_msg = TrackMsg() track_msg.header.stamp = rospy.get_rostime() track_msg.header.frame_id = '/' track_msg.track_type = TrackMsg.TRACK_POSITION_ONLY kin_dict['got_model'] = False for pt in pts.T: p = pt.A1.tolist() pose = Pose(Point(0, 0, 0), Quaternion(0, 0, 0, 1)) pose.position.x = p[0] pose.position.y = p[1] pose.position.z = p[2] track_msg.pose.append(pose) rospy.sleep(0.1) track_pub.publish(track_msg) while kin_dict['got_model'] == False: rospy.sleep(0.01) return kin_dict def init_ros_node(): rospy.init_node('kinematics_estimation_sturm') track_pub = rospy.Publisher('track', TrackMsg) kin_dict = {} track_msg = TrackMsg() rospy.Subscriber('model', ModelMsg, model_cb, kin_dict) return kin_dict, track_pub if __name__ == '__main__': init_ros_node() rospy.spin()
[ [ 1, 0, 0.0238, 0.0119, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0238, 0.0119, 0, 0.66, 0.0769, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0357, 0.0119, 0, 0....
[ "import roslib; roslib.load_manifest('modeling_forces')", "import roslib; roslib.load_manifest('modeling_forces')", "import rospy", "from geometry_msgs.msg import Point32", "from articulation_msgs.msg import ModelMsg", "from articulation_msgs.msg import TrackMsg", "from geometry_msgs.msg import Pose, Po...
import math, numpy as np import glob import roslib; roslib.load_manifest('hrl_tilting_hokuyo') import hrl_tilting_hokuyo.display_3d_mayavi as d3m import matplotlib_util.util as mpu import hrl_lib.util as ut if __name__ == '__main__': import optparse p = optparse.OptionParser() p.add_option('-f', '--fname', action='store', default='', type='string', dest='fname', help='pkl with logged data') p.add_option('-d', '--dir', action='store', default='', type='string', dest='dir', help='directory with logged data') opt, args = p.parse_args() if opt.dir != '': poses_pkl = glob.glob(opt.dir + '/poses_dict*.pkl')[0] d = ut.load_pickle(poses_pkl) elif opt.fname != '': d = ut.load_pickle(opt.fname) else: raise RuntimeError('need either -d or -f') hand_list = d['hand']['pos_list'] hand_rot = d['hand']['rot_list'] if hand_list != []: hand_mat = np.column_stack(hand_list) print 'hand_mat.shape', hand_mat.shape d3m.plot_points(hand_mat, color = (1.,0.,0.), mode='sphere') directions_x = (np.row_stack(hand_rot)[:,0]).T.reshape(len(hand_rot), 3).T directions_z = (np.row_stack(hand_rot)[:,2]).T.reshape(len(hand_rot), 3).T #print 'directions.shape', directions_x.shape #print 'hand_mat.shape', hand_mat.shape #mpu.plot_yx(ut.norm(directions).A1, axis=None) #mpu.show() #mpu.plot_yx(ut.norm(hand_mat[:, 1:] - hand_mat[:, :-1]).A1, axis=None) #mpu.show() d3m.plot_normals(hand_mat, directions_x, color=(1.,0,0.)) d3m.plot_normals(hand_mat, directions_z, color=(0.,0,1.)) mechanism_list = d['mechanism']['pos_list'] mechanism_rot = d['mechanism']['rot_list'] #mechanism_list = [] if mechanism_list != []: mechanism_mat = np.column_stack(mechanism_list) print 'mechanism_mat.shape', mechanism_mat.shape d3m.plot_points(mechanism_mat, color = (0.,0.,1.), mode='sphere') #d3m.plot_points(mechanism_mat, color = (0.,0.,1.), mode='sphere') c = np.row_stack(mechanism_rot)[:,2] directions = c.T.reshape(len(mechanism_rot), 3).T print 'directions.shape', directions.shape print 'mechanism_mat.shape', mechanism_mat.shape d3m.plot_normals(mechanism_mat, directions, color=(0,0,1.)) d3m.show() # mpu.plot_yx(pos_mat[0,:].A1, pos_mat[1,:].A1) # mpu.show()
[ [ 1, 0, 0.0286, 0.0143, 0, 0.66, 0, 526, 0, 2, 0, 0, 526, 0, 0 ], [ 1, 0, 0.0429, 0.0143, 0, 0.66, 0.1429, 958, 0, 1, 0, 0, 958, 0, 0 ], [ 1, 0, 0.0714, 0.0143, 0, ...
[ "import math, numpy as np", "import glob", "import roslib; roslib.load_manifest('hrl_tilting_hokuyo')", "import roslib; roslib.load_manifest('hrl_tilting_hokuyo')", "import hrl_tilting_hokuyo.display_3d_mayavi as d3m", "import matplotlib_util.util as mpu", "import hrl_lib.util as ut", "if __name__ == ...
import commands import os def aggregate_plots(dir, name_list, extension_list): n = len(name_list) for i in range(n): #for nm in name_list: nm = name_list[i] ex = extension_list[i] pngs_list = commands.getoutput('find %s/0* -name "*%s.%s"'%(dir, nm, ex)) pngs_list = pngs_list.splitlines() st = '%s/%s'%(dir, nm) os.system('rm -rf %s'%st) # remove previously existing directory. os.system('mkdir %s'%st) for p in pngs_list: sp = p.split('/') fname = str.join('_',sp[1:]) print 'saving ', fname os.system('cp %s %s'%(p, st+'/'+fname)) if __name__ == '__main__': import optparse p = optparse.OptionParser() p.add_option('-d', '--dir', action='store', default='', type='string', dest='dir', help='directory with logged data') p.add_option('-a', '--agg', action='store_true', dest='agg', help='aggregate plots from different trials into one folder for comparison') opt, args = p.parse_args() if opt.dir == '': raise RuntimeError('Need a directory to work with (-d or --dir)') if opt.agg: name_list = ['mechanism_trajectories_handhook'] extension_list = ['pkl'] # name_list = ['trajectory_check', 'force_components', # 'mechanism_trajectories_handhook', 'poses_dict'] # extension_list = ['png', 'png', 'pkl', 'pkl'] aggregate_plots(opt.dir, name_list, extension_list)
[ [ 1, 0, 0.0426, 0.0213, 0, 0.66, 0, 760, 0, 1, 0, 0, 760, 0, 0 ], [ 1, 0, 0.0638, 0.0213, 0, 0.66, 0.3333, 688, 0, 1, 0, 0, 688, 0, 0 ], [ 2, 0, 0.2979, 0.3617, 0, ...
[ "import commands", "import os", "def aggregate_plots(dir, name_list, extension_list):\n n = len(name_list)\n for i in range(n):\n #for nm in name_list:\n nm = name_list[i]\n ex = extension_list[i]\n pngs_list = commands.getoutput('find %s/0* -name \"*%s.%s\"'%(dir, nm, ex))\n ...
# print 'Following are the key that you can use:' # print '1. Hit ESC to end' # print '2. Hit - to decrease the size of the image' # print '3. Hit + (without the shift key) to increase the image size' # print '4. Hit c to detect chessboard corners' import sys, time import numpy as np, math import glob, commands import roslib; roslib.load_manifest('modeling_forces') import cv from cv_bridge.cv_bridge import CvBridge, CvBridgeError import hrl_lib.util as ut # call func on the 8 neighbors of x,y and on x,y def call_neighborhood(func,func_params,x,y): func(x+0,y+0,*func_params) func(x+1,y+0,*func_params) func(x+1,y-1,*func_params) func(x+0,y-1,*func_params) func(x-1,y-1,*func_params) func(x-1,y+0,*func_params) func(x-1,y+1,*func_params) func(x-0,y+1,*func_params) func(x+1,y+1,*func_params) def on_mouse(event, x, y, flags, param): im, clicked_list = param[0], param[1] scale_factor = param[2] if event == cv.CV_EVENT_LBUTTONDOWN: #call_neighborhood(cv.SetPixel,(im,(255,0,0)),x,y) clicked_list.append((x/scale_factor, y/scale_factor)) # these are in mm def annotate_image(cv_im, mech_info_dict, dir): #cv_im = cv.LoadImage(sys.argv[1], cv.CV_LOAD_IMAGE_GRAYSCALE) size = cv.GetSize(cv_im) print 'Image size:', size[0], size[1] # col, row wnd = 'Image Annotate' cv.NamedWindow(wnd, cv.CV_WINDOW_AUTOSIZE) disp_im = cv.CloneImage(cv_im) new_size = (size[0]/2, size[1]/2) scale_factor = 1 checker_origin_height = mech_info_dict['height'] # chesscoard corners mat cb_dims = (5, 8) # checkerboard dims. (x, y) sq_sz = 19 # size of checkerboard in real units. cb_offset = 500,500 cb_coords = cv.CreateMat(2, cb_dims[0]*cb_dims[1], cv.CV_64FC1) n = 0 for r in range(cb_dims[1]): for c in range(cb_dims[0]): cb_coords[0,n] = c*sq_sz + cb_offset[0] # x coord cb_coords[1,n] = r*sq_sz + cb_offset[1] # y coord n += 1 clicked_list = [] recreate_image = False mechanism_calc_state = 0 mechanism_calc_dict = {} while True: for p in clicked_list: x,y = p[0]*scale_factor, p[1]*scale_factor cv.Circle(disp_im, (x,y), 3, (255,0,0)) cv.ShowImage(wnd, disp_im) k = cv.WaitKey(10) k = k%256 if k == 27: # ESC break elif k == ord('='): # + key without the shift scale_factor = scale_factor*1.2 recreate_image = True elif k == ord('-'): # - key scale_factor = scale_factor/1.2 recreate_image = True elif k == ord('c'): # find chessboard corners. succ, corners = cv.FindChessboardCorners(disp_im, cb_dims) if succ == 0: print 'Chessboard detection FAILED.' else: # chessboard detection was successful cv.DrawChessboardCorners(disp_im, cb_dims, corners, succ) cb_im = cv.CreateMat(2, cb_dims[0]*cb_dims[1], cv.CV_64FC1) corners_mat = np.array(corners).T n = 0 for r in range(cb_dims[1]): for c in range(cb_dims[0]): cb_im[0,n] = corners_mat[0,n] # x coord cb_im[1,n] = corners_mat[1,n] # y coord n += 1 H = cv.CreateMat(3, 3, cv.CV_64FC1) H1 = cv.FindHomography(cb_im, cb_coords, H) Hnp = np.reshape(np.fromstring(H1.tostring()), (3,3)) print 'Homography:' print Hnp d = cv.CloneImage(disp_im) cv.WarpPerspective(d, disp_im, H1, cv.CV_WARP_FILL_OUTLIERS) cv_im = cv.CloneImage(disp_im) elif k == ord('1'): # calculate width of the mechanism del clicked_list[:] cv.SetMouseCallback(wnd, on_mouse, (disp_im, clicked_list, scale_factor)) recreate_image = True print 'Click on two endpoints to mark the width of the mechanism' mechanism_calc_state = 1 elif k == ord('2'): # distance of handle from the hinge del clicked_list[:] cv.SetMouseCallback(wnd, on_mouse, (disp_im, clicked_list, scale_factor)) recreate_image = True print 'Click on handle and hinge to compute distance of handle from hinge.' mechanism_calc_state = 2 elif k == ord('3'): # height of the handle above the ground del clicked_list[:] cv.SetMouseCallback(wnd, on_mouse, (disp_im, clicked_list, scale_factor)) recreate_image = True print 'Click on handle top and bottom to compute height of handle above the ground.' mechanism_calc_state = 3 elif k == ord('4'): # top and bottom edge of the mechanism del clicked_list[:] cv.SetMouseCallback(wnd, on_mouse, (disp_im, clicked_list, scale_factor)) recreate_image = True print 'Click on top and bottom edges of the mechanism.' mechanism_calc_state = 4 elif k ==ord('d'): # display the calculated values print 'mechanism_calc_dict:', mechanism_calc_dict print 'mech_info_dict:', mech_info_dict elif k == ord('s'): # save the pkl ut.save_pickle(mechanism_calc_dict, dir+'/mechanism_calc_dict.pkl') print 'Saved the pickle' #elif k != -1: elif k != 255: print 'k:', k if recreate_image: new_size = (int(size[0]*scale_factor), int(size[1]*scale_factor)) disp_im = cv.CreateImage(new_size, cv_im.depth, cv_im.nChannels) cv.Resize(cv_im, disp_im) cv.SetMouseCallback(wnd, on_mouse, (disp_im, clicked_list, scale_factor)) recreate_image = False if len(clicked_list) == 2: if mechanism_calc_state == 1: w = abs(clicked_list[0][0] - clicked_list[1][0]) print 'Width in mm:', w mechanism_calc_dict['mech_width'] = w/1000. if mechanism_calc_state == 2: w = abs(clicked_list[0][0] - clicked_list[1][0]) print 'Width in mm:', w mechanism_calc_dict['handle_hinge_dist'] = w/1000. if mechanism_calc_state == 3: p1, p2 = clicked_list[0], clicked_list[1] h1 = (cb_offset[1] - p1[1])/1000. + checker_origin_height h2 = (cb_offset[1] - p2[1])/1000. + checker_origin_height mechanism_calc_dict['handle_top'] = max(h1, h2) mechanism_calc_dict['handle_bottom'] = min(h1, h2) if mechanism_calc_state == 4: p1, p2 = clicked_list[0], clicked_list[1] h1 = (cb_offset[1] - p1[1])/1000. + checker_origin_height h2 = (cb_offset[1] - p2[1])/1000. + checker_origin_height mechanism_calc_dict['mechanism_top'] = max(h1, h2) mechanism_calc_dict['mechanism_bottom'] = min(h1, h2) mechanism_calc_state = 0 if __name__ == '__main__': import optparse p = optparse.OptionParser() p.add_option('-d', '--dir', action='store', default='', type='string', dest='dir', help='mechanism directory') opt, args = p.parse_args() # dir_list = commands.getoutput('ls -d %s/*/'%(opt.dir)).splitlines() # dir_list = dir_list[:] # for direc in dir_list: img_list = glob.glob(opt.dir+'/*.jpg') mech_info_dict = ut.load_pickle(opt.dir+'/mechanism_info.pkl') for img in img_list: cv_im = cv.LoadImage(img) annotate_image(cv_im, mech_info_dict, opt.dir) sys.exit() # cv.DestroyAllWindows() #print 'k:', chr(k)
[ [ 1, 0, 0.034, 0.0049, 0, 0.66, 0, 509, 0, 2, 0, 0, 509, 0, 0 ], [ 1, 0, 0.0388, 0.0049, 0, 0.66, 0.0909, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0437, 0.0049, 0, 0...
[ "import sys, time", "import numpy as np, math", "import glob, commands", "import roslib; roslib.load_manifest('modeling_forces')", "import roslib; roslib.load_manifest('modeling_forces')", "import cv", "from cv_bridge.cv_bridge import CvBridge, CvBridgeError", "import hrl_lib.util as ut", "def call_...
import csv import numpy as np import pylab as pb import matplotlib_util.util as mpu def load_data(): mk = csv.reader(open('Mechanism Kinematics.csv', 'U')) data = [] for r in mk: data.append(r) return data ## # Deal with repetition def expand(data, keys): repetition_idx = np.where(np.array(keys) == 'repetition')[0] data[repetition_idx] = [int(e) for e in data[repetition_idx]] repeated_data = [] for i in range(len(data)): repeated_data.append([]) for current_instance in range(len(data[0])): if data[repetition_idx][current_instance] > 1: for rep in range(1, data[repetition_idx][current_instance]): for l, rl in zip(data, repeated_data): rl.append(l[current_instance]) for l, rl in zip(data, repeated_data): l += rl data.pop(repetition_idx) keys.pop(repetition_idx) return data def test_expand(): data = [['a', 'b', 'c'], ['A', 'B', 'C'], [1, 4, 10]] keys = ['letters', 'LETTERS', 'repetition'] new_data = expand(data, keys) print new_data def extract_keys(csv_data, keys): format = csv_data[1] indices = [] llists = [] for k in keys: indices.append(np.where(np.array(format) == k)[0]) llists.append([]) for i in range(2, len(csv_data)): if len(csv_data[i]) == 0: break if len(csv_data[i][indices[0]]) > 0: for lidx, data_idx in enumerate(indices): llists[lidx].append(csv_data[i][data_idx]) return llists def plot_radii(csv_data, color='#3366FF'): keys = ['radius', 'type', 'name', 'repetition'] llists = expand(extract_keys(csv_data, keys), keys) rad_list = np.array([float(r) for r in llists[0]]) / 100.0 types = llists[1] names = np.array(llists[2]) all_types = set(types) print 'Radii types', all_types types_arr = np.array(types) # np.where(types_arr == 'C') cabinet_rad_list = rad_list[np.where(types_arr == 'C')[0]] others_rad_list = rad_list[np.where(types_arr != 'C')[0]] other_names = names[np.where(types_arr != 'C')[0]] print '>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>' print 'radii other names' for i, n in enumerate(other_names): print n, others_rad_list[i] print '>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>' rad_lists = [rad_list, cabinet_rad_list, others_rad_list] titles = ['Radii of Rotary Mechanisms', 'Radii of Cabinets', 'Radii of Other Mechanisms'] bin_width = 0.05 max_radius = np.max(rad_list) print 'MIN RADIUS', np.min(rad_list) print 'MAX RADIUS', max_radius mpu.set_figure_size(5.,5.) for idx, radii in enumerate(rad_lists): f = pb.figure() f.set_facecolor('w') bins = np.arange(0.-bin_width/2., max_radius+2*bin_width, bin_width) hist, bin_edges = np.histogram(radii, bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Radius (meters)', ylabel='\# of mechanisms', plot_title=titles[idx], color=color, label='All') pb.xlim(.1, 1.) pb.ylim(0, 55) mpu.legend(display_mode = 'less_space', handlelength=1.) #-- different classes in different colors in the same histogram. f = pb.figure() f.set_facecolor('w') bins = np.arange(0.-bin_width/2., max_radius+2*bin_width, bin_width) hist, bin_edges = np.histogram(rad_lists[0], bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Radius (meters)', ylabel='\# of mechanisms', plot_title=titles[1], color='g', label='Cabinets') hist, bin_edges = np.histogram(rad_lists[2], bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Radius (meters)', ylabel='\# of mechanisms', plot_title='Cabinets and Other Mechanisms', color='y', label='Other') pb.xlim(.1, 1.) pb.ylim(0, 55) mpu.legend(display_mode = 'less_space', handlelength=1.) color_list = ['g', 'b', 'r'] marker_list = ['s', '^', 'v'] label_list = ['All', 'Cabinets', 'Other'] scatter_size_list = [8, 5, 5] mpu.set_figure_size(5.,5.) mpu.figure() for idx, radii in enumerate(rad_lists): bins = np.arange(0.-bin_width/2., max_radius+2*bin_width, bin_width) hist, bin_edges = np.histogram(radii, bins) bin_midpoints = np.arange(0., max_radius+bin_width, bin_width) mpu.plot_yx(hist, bin_midpoints, color = color_list[idx], alpha = 0.6, marker = marker_list[idx], scatter_size = scatter_size_list[idx], xlabel='Radius (meters)', ylabel='\# of mechanisms', label = label_list[idx]) mpu.legend(display_mode = 'less_space') def plot_opening(csv_data): keys = ['distance', 'type', 'repetition'] llists = expand(extract_keys(csv_data, keys), keys) dists = np.array([float(f) for f in llists[0]])/100. types = llists[1] types_arr = np.array(types) drawer_dists = dists[np.where(types_arr == 'R')[0]] other_dists = dists[np.where(types_arr != 'R')[0]] print 'Opening distances types', set(types) bin_width = 0.02 bins = np.arange(-bin_width/2., np.max(dists)+2*bin_width, bin_width) dists_list = [dists]#, drawer_dists, other_dists] titles = ['Opening Distances of Drawers']#, 'drawer', 'other'] # print 'Total number of drawers:', len(dists) mpu.set_figure_size(5.,4.) for idx, d in enumerate(dists_list): f = pb.figure() f.set_facecolor('w') hist, bin_edges = np.histogram(d, bins) #import pdb; pdb.set_trace() mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=0.8*bin_width, xlabel='Opening Distance (meters)', ylabel='\# of Mechanisms', plot_title=titles[idx], color='#3366FF') # pb.xticks(bins[np.where(bins > np.min(dists))[0][0]-2:-1]) # pb.yticks(range(0, 26, 5)) # pb.ylim(0, 25) def handle_height_histogram(mean_height_list, plot_title='', color='#3366FF', max_height=2.5, bin_width=.1, ymax=35): bins = np.arange(0.-bin_width/2., max_height+2*bin_width, bin_width) hist, bin_edges = np.histogram(np.array(mean_height_list), bins) f = pb.figure() f.set_facecolor('w') f.subplots_adjust(bottom=.16, top=.86) mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=bin_width*0.8, plot_title=plot_title, xlabel='Height (meters)', ylabel='\# of mechanisms', color=color) pb.xlim(0, max_height) pb.ylim(0, ymax) def handle_height_histogram_advait(mean_height_list, plot_title='', color='#3366FF', max_height=2.2, bin_width=.1, ymax=35, new_figure = True, label = '__no_legend__'): if new_figure: mpu.set_figure_size(5.,4.) # f = mpu.figure() # f.subplots_adjust(bottom=.25, top=.99, right=0.99, left=0.12) bins = np.arange(0.-bin_width/2., max_height+2*bin_width, bin_width) hist, bin_edges = np.histogram(np.array(mean_height_list), bins) h = mpu.plot_histogram(bin_edges[:-1]+bin_width/2., hist, width=bin_width*0.8, plot_title=plot_title, xlabel='Height (meters)', ylabel='\# of mechanisms', color=color, label = label) pb.xlim(0, max_height) pb.ylim(0, ymax) return h def plot_rotary_heights(csv_data): keys = ['radius', 'bottom', 'top', 'type', 'name', 'repetition'] llists = expand(extract_keys(csv_data, keys), keys) types = llists[3] names = np.array(llists[4]) print 'Handle bottom edge types', set(types) bottom_pts = np.array([float(f) for f in llists[1]])/100. top_pts = [] for i,f in enumerate(llists[2]): if f == '': f = llists[1][i] top_pts.append(float(f)/100.) top_pts = np.array(top_pts) print 'total number of doors:', len(top_pts) mid_pts = (bottom_pts + top_pts)/2. types_arr = np.array(types) cabinet_mid_pts = mid_pts[np.where(types_arr == 'C')[0]] other_mid_pts = mid_pts[np.where(types_arr != 'C')[0]] other_names = names[np.where(types_arr != 'C')[0]] print 'other names', other_names # Hai, Advait apologizes for changing code without making a copy. # He didn't realise. He'll make a copy from an earlier revision in # subversion soon. ymax = 85 handle_height_histogram_advait(mid_pts, plot_title='Handle Heights', ymax=ymax, label = 'All') mpu.legend(display_mode = 'less_space', handlelength=1.) handle_height_histogram_advait(mid_pts, plot_title='Handle Heights', ymax=ymax, color = 'g', label = 'Cabinets') handle_height_histogram_advait(other_mid_pts, plot_title='Handle Heights', ymax=ymax, color = 'y', new_figure = False, label = 'Other') mpu.legend(display_mode = 'less_space', handlelength=1.) def plot_prismatic_heights(csv_data): keys = ['distance', 'bottom', 'top', 'type', 'name', 'home', 'repetition'] llists = expand(extract_keys(csv_data, keys), keys) types = llists[3] names = np.array(llists[4]) home = np.array(llists[5]) print 'Handle bottom edge types', set(types) bottom_pts = np.array([float(f) for f in llists[1]])/100. top_pts = [] for i,f in enumerate(llists[2]): if f == '': f = llists[1][i] top_pts.append(float(f)/100.) top_pts = np.array(top_pts) mid_pts = (bottom_pts + top_pts)/2. types_arr = np.array(types) max_height = np.max(bottom_pts) sort_order = np.argsort(bottom_pts) names = names[sort_order] bottom_pts = bottom_pts[sort_order] home = home[sort_order] for i, name in enumerate(names): print home[i], name, bottom_pts[i] handle_height_histogram(bottom_pts, plot_title='Heights of Handle Lower Edge (Drawers)', max_height = max_height) max_height = np.max(mid_pts) handle_height_histogram_advait(mid_pts, plot_title='Handle Heights', max_height = max_height+0.1, ymax=43) #test_expand() csv_data = load_data() plot_prismatic_heights(csv_data) #plot_radii(csv_data) plot_rotary_heights(csv_data) #plot_opening(csv_data) pb.show()
[ [ 1, 0, 0.0037, 0.0037, 0, 0.66, 0, 312, 0, 1, 0, 0, 312, 0, 0 ], [ 1, 0, 0.0074, 0.0037, 0, 0.66, 0.0588, 954, 0, 1, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0112, 0.0037, 0, ...
[ "import csv", "import numpy as np", "import pylab as pb", "import matplotlib_util.util as mpu", "def load_data():\n mk = csv.reader(open('Mechanism Kinematics.csv', 'U'))\n data = []\n for r in mk:\n data.append(r)\n\n return data", " mk = csv.reader(open('Mechanism Kinematics.csv', ...
#!/usr/bin/python import roslib; roslib.load_manifest('modeling_forces') import rospy import force_torque.FTClient as ftc import hrl_lib.util as ut from std_msgs.msg import Empty import time import numpy as np import glob ## # make an appropriate plot. # @param d - dictionary saved in the pkl def plot_ft(d): ft_list = d['ft_list'] time_list = d['time_list'] ft_mat = np.matrix(ft_list).T # 6xN np matrix force_mat = ft_mat[0:3, :] tarray = np.array(time_list) print 'average rate', 1. / np.mean(tarray[1:] - tarray[:-1]) time_list = (tarray - tarray[0]).tolist() print len(time_list) force_mag_l = ut.norm(force_mat).A1.tolist() #for i,f in enumerate(force_mag_l): # if f > 15: # break #force_mag_l = force_mag_l[i:] #time_list = time_list[i:] mpu.plot_yx(force_mag_l, time_list, axis=None, label='Force Magnitude', xlabel='Time(seconds)', ylabel='Force Magnitude (N)', color = mpu.random_color()) def got_trigger_cb(data, d): d['flag'] = True if __name__ == '__main__': import optparse p = optparse.OptionParser() p.add_option('-l', '--log', action='store_true', dest='log', help='log FT data') p.add_option('-p', '--plot', action='store_true', dest='plot', help='plot FT data') p.add_option('-r', '--ros', action='store_true', dest='ros', help='start and stop logging messages over ROS.') p.add_option('-a', '--all', action='store_true', dest='all', help='plot all the pkls in a single plot') p.add_option('-f', '--fname', action='store', default='', type='string', dest='fname', help='pkl with logged data') opt, args = p.parse_args() if opt.log: client = ftc.FTClient('/force_torque_ft2') l = [] time_list = [] if opt.ros: topic_name_cb = '/ftlogger/trigger' got_trigger_dict = {'flag': False} rospy.Subscriber(topic_name_cb, Empty, got_trigger_cb, got_trigger_dict) while got_trigger_dict['flag'] == False: time.sleep(0.05) got_trigger_dict['flag'] = False while not rospy.is_shutdown(): ft, t_msg = client.read(fresh=True, with_time_stamp=True) if ft != None: l.append(ft.A1.tolist()) time_list.append(t_msg) if got_trigger_dict['flag'] == True: break time.sleep(1/1000.0) else: print 'Sleeping for 5 seconds.' time.sleep(5.) print '>>>>>>>>>>>>>>>>>>>>>>>>>>>>' print 'BEGIN' print '>>>>>>>>>>>>>>>>>>>>>>>>>>>>' client.bias() t_start = time.time() t_now = time.time() logging_time = 5. while (t_now - t_start) < logging_time: ft, t_msg = client.read(fresh=True, with_time_stamp=True) t_now = time.time() if ft != None: l.append(ft.A1.tolist()) time_list.append(t_msg) time.sleep(1/1000.0) print 'saving the pickle' d = {} d['ft_list'] = l d['time_list'] = time_list fname = 'ft_log_'+ut.formatted_time()+'.pkl' ut.save_pickle(d, fname) if opt.plot: import matplotlib_util.util as mpu if opt.fname == '' and (not opt.all): raise RuntimeError('need a pkl name (-f or --fname) or --all') if opt.all: fname_list = glob.glob('ft_log*.pkl') else: fname_list = [opt.fname] for fname in fname_list: d = ut.load_pickle(fname) plot_ft(d) mpu.legend() mpu.show()
[ [ 1, 0, 0.0233, 0.0078, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0233, 0.0078, 0, 0.66, 0.0909, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.031, 0.0078, 0, 0.6...
[ "import roslib; roslib.load_manifest('modeling_forces')", "import roslib; roslib.load_manifest('modeling_forces')", "import rospy", "import force_torque.FTClient as ftc", "import hrl_lib.util as ut", "from std_msgs.msg import Empty", "import time", "import numpy as np", "import glob", "def plot_ft...
import commands import sys, os import glob l1 = glob.glob('data_1tb/*/mechanism_info.pkl') l2 = [] for d in l1: l2.append('/'.join(['aggregated_pkls_Feb11']+d.split('/')[1:])) for d1,d2 in zip(l1,l2): os.system('cp %s %s'%(d1, d2))
[ [ 1, 0, 0.1538, 0.0769, 0, 0.66, 0, 760, 0, 1, 0, 0, 760, 0, 0 ], [ 1, 0, 0.2308, 0.0769, 0, 0.66, 0.1667, 509, 0, 2, 0, 0, 509, 0, 0 ], [ 1, 0, 0.3077, 0.0769, 0, ...
[ "import commands", "import sys, os", "import glob", "l1 = glob.glob('data_1tb/*/mechanism_info.pkl')", "l2 = []", "for d in l1:\n l2.append('/'.join(['aggregated_pkls_Feb11']+d.split('/')[1:]))", " l2.append('/'.join(['aggregated_pkls_Feb11']+d.split('/')[1:]))", "for d1,d2 in zip(l1,l2):\n o...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # Author: Advait Jain import roslib; roslib.load_manifest('2009_humanoids_epc_pull') import scipy.optimize as so import math, numpy as np import pylab as pl import sys, optparse, time import copy from enthought.mayavi import mlab import mekabot.hrl_robot as hr #import util as ut import hrl_lib.util as ut, hrl_lib.transforms as tr import matplotlib_util.util as mpu import hrl_tilting_hokuyo.display_3d_mayavi as d3m #import shapely.geometry as sg class JointTrajectory(): ''' class to store joint trajectories. data only - use for pickling. ''' def __init__(self): self.time_list = [] # time in seconds self.q_list = [] #each element is a list of 7 joint angles. self.qdot_list = [] #each element is a list of 7 joint angles. class CartesianTajectory(): ''' class to store trajectory of cartesian points. data only - use for pickling ''' def __init__(self): self.time_list = [] # time in seconds self.p_list = [] #each element is a list of 3 coordinates class ForceTrajectory(): ''' class to store time evolution of the force at the end effector. data only - use for pickling ''' def __init__(self): self.time_list = [] # time in seconds self.f_list = [] #each element is a list of 3 coordinates def joint_to_cartesian(traj): ''' traj - JointTrajectory returns CartesianTajectory after performing FK on traj. ''' firenze = hr.M3HrlRobot(connect=False) pts = [] for q in traj.q_list: p = firenze.FK('right_arm',q) pts.append(p.A1.tolist()) ct = CartesianTajectory() ct.time_list = copy.copy(traj.time_list) ct.p_list = copy.copy(pts) #return np.matrix(pts).T return ct def plot_forces_quiver(pos_traj,force_traj,color='k'): import arm_trajectories as at #if traj.__class__ == at.JointTrajectory: if isinstance(pos_traj,at.JointTrajectory): pos_traj = joint_to_cartesian(pos_traj) pts = np.matrix(pos_traj.p_list).T label_list = ['X coord (m)', 'Y coord (m)', 'Z coord (m)'] x = pts[0,:].A1.tolist() y = pts[1,:].A1.tolist() forces = np.matrix(force_traj.f_list).T u = (-1*forces[0,:]).A1.tolist() v = (-1*forces[1,:]).A1.tolist() pl.quiver(x,y,u,v,width=0.002,color=color,scale=100.0) # pl.quiver(x,y,u,v,width=0.002,color=color) pl.axis('equal') def plot_cartesian(traj,xaxis=None,yaxis=None,zaxis=None,color='b',label='_nolegend_', linewidth=2,scatter_size=20): ''' xaxis - x axis for the graph (0,1 or 2) zaxis - for a 3d plot. not implemented. ''' import arm_trajectories as at #if traj.__class__ == at.JointTrajectory: if isinstance(traj,at.JointTrajectory): traj = joint_to_cartesian(traj) pts = np.matrix(traj.p_list).T label_list = ['X coord (m)', 'Y coord (m)', 'Z coord (m)'] x = pts[xaxis,:].A1.tolist() y = pts[yaxis,:].A1.tolist() if zaxis == None: pl.plot(x,y,c=color,linewidth=linewidth,label=label) pl.scatter(x,y,c=color,s=scatter_size,label='_nolegend_', linewidths=0) pl.xlabel(label_list[xaxis]) pl.ylabel(label_list[yaxis]) pl.legend(loc='best') pl.axis('equal') else: from numpy import array from enthought.mayavi.api import Engine engine = Engine() engine.start() if len(engine.scenes) == 0: engine.new_scene() z = pts[zaxis,:].A1.tolist() time_list = [t-traj.time_list[0] for t in traj.time_list] mlab.plot3d(x,y,z,time_list,tube_radius=None,line_width=4) mlab.axes() mlab.xlabel(label_list[xaxis]) mlab.ylabel(label_list[yaxis]) mlab.zlabel(label_list[zaxis]) mlab.colorbar(title='Time') # ------------------------------------------- axes = engine.scenes[0].children[0].children[0].children[1] axes.axes.position = array([ 0., 0.]) axes.axes.label_format = '%-#6.2g' axes.title_text_property.font_size=4 ## compute the force that the arm would apply given the stiffness matrix # @param q_actual_traj - Joint Trajectory (actual angles.) # @param q_eq_traj - Joint Trajectory (equilibrium point angles.) # @param torque_traj - JointTrajectory (torques measured at the joints.) # @param rel_stiffness_list - list of 5 elements (stiffness numbers for the joints.) # @return lots of things, look at the code. def compute_forces(q_actual_traj,q_eq_traj,torque_traj,rel_stiffness_list): firenze = hr.M3HrlRobot(connect=False) d_gains_list_mN_deg_sec = [-100,-120,-15,-25,-1.25] d_gains_list = [180./1000.*s/math.pi for s in d_gains_list_mN_deg_sec] stiff_list_mNm_deg = [1800,1300,350,600,60] stiff_list_Nm_rad = [180./1000.*s/math.pi for s in stiff_list_mNm_deg] # stiffness_settings = [0.15,0.7,0.8,0.8,0.8] # dia = np.array(stiffness_settings) * np.array(stiff_list_Nm_rad) dia = np.array(rel_stiffness_list) * np.array(stiff_list_Nm_rad) k_q = np.matrix(np.diag(dia)) dia_inv = 1./dia k_q_inv = np.matrix(np.diag(dia_inv)) actual_cart = joint_to_cartesian(q_actual_traj) eq_cart = joint_to_cartesian(q_eq_traj) force_traj_jacinv = ForceTrajectory() force_traj_stiff = ForceTrajectory() force_traj_torque = ForceTrajectory() k_cart_list = [] for q_actual,q_dot,q_eq,actual_pos,eq_pos,t,tau_m in zip(q_actual_traj.q_list,q_actual_traj.qdot_list,q_eq_traj.q_list,actual_cart.p_list,eq_cart.p_list,q_actual_traj.time_list,torque_traj.q_list): q_eq = firenze.clamp_to_physical_joint_limits('right_arm',q_eq) q_delta = np.matrix(q_actual).T - np.matrix(q_eq).T tau = k_q * q_delta[0:5,0] - np.matrix(np.array(d_gains_list)*np.array(q_dot)[0:5]).T x_delta = np.matrix(actual_pos).T - np.matrix(eq_pos).T jac_full = firenze.Jac('right_arm',q_actual) jac = jac_full[0:3,0:5] jac_full_eq = firenze.Jac('right_arm',q_eq) jac_eq = jac_full_eq[0:3,0:5] k_cart = np.linalg.inv((jac_eq*k_q_inv*jac_eq.T)) # calculating stiff matrix using Jacobian for eq pt. k_cart_list.append(k_cart) pseudo_inv_jac = np.linalg.inv(jac_full*jac_full.T)*jac_full tau_full = np.row_stack((tau,np.matrix(tau_m[5:7]).T)) #force = (-1*pseudo_inv_jac*tau_full)[0:3] force = -1*pseudo_inv_jac[0:3,0:5]*tau force_traj_jacinv.f_list.append(force.A1.tolist()) force_traj_stiff.f_list.append((k_cart*x_delta).A1.tolist()) force_traj_torque.f_list.append((pseudo_inv_jac*np.matrix(tau_m).T)[0:3].A1.tolist()) return force_traj_jacinv,force_traj_stiff,force_traj_torque,k_cart_list ## return two lists containing the radial and tangential components of the forces. # @param f_list - list of forces. (each force is a list of 2 or 3 floats) # @param p_list - list of positions. (each position is a list of 2 or 3 floats) # @param cx - x coord of the center of the circle. # @param cy - y coord of the center of the circle. # @return list of magnitude of radial component, list of magnitude tangential component. def compute_radial_tangential_forces(f_list,p_list,cx,cy): f_rad_l,f_tan_l = [],[] for f,p in zip(f_list,p_list): rad_vec = np.array([p[0]-cx,p[1]-cy]) rad_vec = rad_vec/np.linalg.norm(rad_vec) f_vec = np.array([f[0],f[1]]) f_rad_mag = np.dot(f_vec,rad_vec) f_tan_mag = np.linalg.norm(f_vec-rad_vec*f_rad_mag) f_rad_mag = abs(f_rad_mag) f_rad_l.append(f_rad_mag) f_tan_l.append(f_tan_mag) return f_rad_l,f_tan_l def plot_error_forces(measured_list,calc_list): err_x, err_y = [],[] err_rel_x, err_rel_y = [],[] mag_x, mag_y = [],[] for m,c in zip(measured_list,calc_list): err_x.append(abs(m[0]-c[0])) err_rel_x.append(abs(m[0]-c[0])/abs(m[0])*100) #err_rel_x.append(ut.bound(abs(m[0]-c[0])/abs(m[0])*100,100,0)) mag_x.append(abs(m[0])) err_y.append(abs(m[1]-c[1])) err_rel_y.append(abs(m[1]-c[1])/abs(m[1])*100) #err_rel_y.append(ut.bound(abs(m[1]-c[1])/abs(m[1])*100,100,0)) mag_y.append(abs(m[1])) x_idx = range(len(err_x)) zero = [0 for i in x_idx] fig = pl.figure() ax1 = fig.add_subplot(111) ax2 = ax1.twinx() ax1.plot(zero,c='k',linewidth=1,label='_nolegend_') l1 = ax1.plot(err_x,c='b',linewidth=1,label='absolute error') ax1.scatter(x_idx,err_x,c='b',s=10,label='_nolegend_', linewidths=0) l2 = ax1.plot(mag_x,c='g',linewidth=1,label='magnitude') ax1.scatter(x_idx,mag_x,c='g',s=10,label='_nolegend_', linewidths=0) l3 = ax2.plot(err_rel_x,c='r',linewidth=1,label='relative error %') ax1.set_ylim(0,15) ax2.set_ylim(0,100) ax1.set_xlabel('measurement number') ax1.set_ylabel('x component of force (N)') ax2.set_ylabel('percentage error') ax1.yaxis.set_label_coords(-0.3,0.5) ax2.yaxis.set_label_coords(-0.3,0.5) leg = pl.legend([l1,l2,l3],['absolute error','magnitude','rel error %'],loc='upper left', handletextsep=0.015,handlelen=0.003,labelspacing=0.003) fig = pl.figure() ax1 = fig.add_subplot(111) ax2 = ax1.twinx() ax1.plot(zero,c='k',linewidth=1) l1 = ax1.plot(err_y,c='b',linewidth=1) ax1.scatter(x_idx,err_y,c='b',s=10, linewidths=0) l2 = ax1.plot(mag_y,c='g',linewidth=1) ax1.scatter(x_idx,mag_y,c='g',s=10,linewidths=0) l3 = ax2.plot(err_rel_y,c='r',linewidth=1) ax1.set_ylim(0,15) ax2.set_ylim(0,100) ax1.yaxis.set_label_coords(-0.3,0.5) ax2.yaxis.set_label_coords(-0.3,0.5) ax1.set_xlabel('measurement number') ax1.set_ylabel('y component of force (N)') ax2.set_ylabel('percentage error') #pl.legend(loc='best') leg = pl.legend([l1,l2,l3],['absolute error','magnitude','rel error %'],loc='upper left', handletextsep=0.015,handlelen=0.003,labelspacing=0.003) # pl.figure() # pl.plot(zero,c='k',linewidth=0.5,label='_nolegend_') # pl.plot(err_y,c='b',linewidth=1,label='error') # pl.plot(err_rel_y,c='r',linewidth=1,label='relative error %') # pl.scatter(x_idx,err_y,c='b',s=10,label='_nolegend_', linewidths=0) # pl.plot(mag_y,c='g',linewidth=1,label='magnitude') # pl.scatter(x_idx,mag_y,c='g',s=10,label='_nolegend_', linewidths=0) # # pl.xlabel('measurement number') # pl.ylabel('y component of force (N)') # pl.legend(loc='best') # pl.axis('equal') def plot_stiff_ellipses(k_cart_list,pos_traj,skip=10,subplotnum=111): import arm_trajectories as at if isinstance(pos_traj,at.JointTrajectory): pos_traj = joint_to_cartesian(pos_traj) pts = np.matrix(pos_traj.p_list).T x_l = pts[0,:].A1.tolist() y_l = pts[1,:].A1.tolist() from pylab import figure, show, rand from matplotlib.patches import Ellipse ells = [] scale = 25000. ratio_list = [] for k_c,x,y in zip(k_cart_list[::skip],x_l[::skip],y_l[::skip]): w,v = np.linalg.eig(k_c[0:2,0:2]) w_abs = np.abs(w) major_axis = np.max(w_abs) minor_axis = np.min(w_abs) print 'major, minor:',major_axis,minor_axis # print 'k_c:', k_c ratio_list.append(major_axis/minor_axis) ells.append(Ellipse(np.array([x,y]),width=w[0]/scale,height=w[1]/scale,angle=math.degrees(math.atan2(v[1,0],v[0,0])))) ells[-1].set_lw(2) #fig = pl.figure() #ax = fig.add_subplot(111, aspect='equal') ax = pl.subplot(subplotnum, aspect='equal') for e in ells: ax.add_artist(e) #e.set_clip_box(ax.bbox) #e.set_alpha(1.) e.set_facecolor(np.array([1,1,1])) plot_cartesian(pos_traj,xaxis=0,yaxis=1,color='b', linewidth=0.0,scatter_size=0) # plot_cartesian(pos_traj,xaxis=0,yaxis=1,color='b',label='Eq Point', # linewidth=1.5,scatter_size=0) # plot_cartesian(d['actual'],xaxis=0,yaxis=1,color='b',label='FK', # linewidth=1.5,scatter_size=0) # plot_cartesian(d['eq_pt'], xaxis=0,yaxis=1,color='g',label='Eq Point', # linewidth=1.5,scatter_size=0) mean_ratio = np.mean(np.array(ratio_list)) std_ratio = np.std(np.array(ratio_list)) return mean_ratio,std_ratio # plot the force field in the xy plane for the stiffness matrix k_cart. ## @param k_cart: 3x3 cartesian space stiffness matrix. def plot_stiffness_field(k_cart,plottitle=''): n_points = 20 ang_step = math.radians(360)/n_points x_list = [] y_list = [] u_list = [] v_list = [] k_cart = k_cart[0:2,0:2] for i in range(n_points): ang = i*ang_step for r in [0.5,1.,1.5]: dx = r*math.cos(ang) dy = r*math.sin(ang) dtau = -k_cart*np.matrix([dx,dy]).T x_list.append(dx) y_list.append(dy) u_list.append(dtau[0,0]) v_list.append(dtau[1,0]) pl.figure() # mpu.plot_circle(0,0,1.0,0.,math.radians(360)) mpu.plot_radii(0,0,1.5,0.,math.radians(360),interval=ang_step,color='r') pl.quiver(x_list,y_list,u_list,v_list,width=0.002,color='k',scale=None) pl.axis('equal') pl.title(plottitle) def plot_stiff_ellipse_map(stiffness_list,num): firenze = hr.M3HrlRobot(connect=False) hook_3dprint_angle = math.radians(20-2.54) rot_mat = tr.Rz(0.-hook_3dprint_angle)*tr.Ry(math.radians(-90)) d_gains_list_mN_deg_sec = [-100,-120,-15,-25,-1.25] d_gains_list = [180./1000.*s/math.pi for s in d_gains_list_mN_deg_sec] stiff_list_mNm_deg = [1800,1300,350,600,60] stiff_list_Nm_rad = [180./1000.*s/math.pi for s in stiff_list_mNm_deg] dia = np.array(stiffness_list) * np.array(stiff_list_Nm_rad) k_q = np.matrix(np.diag(dia)) dia_inv = 1./dia k_q_inv = np.matrix(np.diag(dia_inv)) s0,s1,s2,s3 = stiffness_list[0],stiffness_list[1],stiffness_list[2],stiffness_list[3] i=0 #for z in np.arange(-0.1,-0.36,-0.05): for z in np.arange(-0.23,-0.27,-0.05): pl.figure() k_cart_list = [] pos_traj = CartesianTajectory() for x in np.arange(0.25,0.56,0.05): for y in np.arange(-0.15,-0.56,-0.05): if math.sqrt(x**2+y**2)>0.55: continue q = firenze.IK('right_arm',np.matrix([x,y,z]).T,rot_mat) if q == None: continue jac_full = firenze.Jac('right_arm',q) jac = jac_full[0:3,0:5] k_cart = np.linalg.inv((jac*k_q_inv*jac.T)) k_cart_list.append(k_cart) pos_traj.p_list.append([x,y,z]) pos_traj.time_list.append(0.1) if len(pos_traj.p_list)>0: ret = plot_stiff_ellipses(k_cart_list,pos_traj,skip=1) pl.axis('equal') pl.legend(loc='best') title_string = 'z: %.2f stiff:[%.1f,%.1f,%.1f,%.1f]'%(z,s0,s1,s2,s3) pl.title(title_string) i+=1 pl.savefig('ellipses_%03d_%03d.png'%(num,i)) return ret def compute_workspace(z,plot=False,wrist_roll_angle=math.radians(0),subplotnum=None,title=''): firenze = hr.M3HrlRobot(connect=False) # hook_3dprint_angle = math.radians(20-2.54) # rot_mat = tr.Rz(math.radians(-90.)-hook_3dprint_angle)*tr.Ry(math.radians(-90)) rot_mat = tr.Rz(wrist_roll_angle)*tr.Ry(math.radians(-90)) x_list,y_list = [],[] for x in np.arange(0.15,0.65,0.02): for y in np.arange(-0.05,-0.65,-0.02): q = firenze.IK('right_arm',np.matrix([x,y,z]).T,rot_mat) if q != None: x_list.append(x) y_list.append(y) if len(x_list) > 2: multipoint = sg.Point(x_list[0],y_list[0]) for x,y in zip(x_list[1:],y_list[1:]): multipoint = multipoint.union(sg.Point(x,y)) hull = multipoint.convex_hull if plot: coords_seq = hull.boundary.coords hull_x_list,hull_y_list = [],[] for c in coords_seq: hull_x_list.append(c[0]) hull_y_list.append(c[1]) mpu.plot_yx(y_list,x_list,linewidth=0,subplotnum=subplotnum,axis='equal', plot_title=title) mpu.plot_yx(hull_y_list,hull_x_list,linewidth=2,subplotnum=subplotnum,axis='equal') return hull,len(x_list) else: return None,None def diff_roll_angles(): pl.subplot(211,aspect='equal') # search along z coord and make a histogram of the areas def compute_workspace_z(): n_points_list,area_list,z_list = [],[],[] #for z in np.arange(-0.1,-0.36,-0.02): #for z in np.arange(-0.05,-0.35,-0.01): for z in np.arange(-0.15,-0.155,-0.01): pl.figure() hull,n_points = compute_workspace(z,plot=True) pl.title('z: %.2f'%(z)) pl.savefig('z_%.2f.png'%(z)) # hull,n_points = compute_workspace(z,plot=False) if hull != None: area_list.append(hull.area) z_list.append(z) n_points_list.append(n_points) coords_seq = hull.boundary.coords hull_x_list,hull_y_list = [],[] for c in coords_seq: hull_x_list.append(c[0]) hull_y_list.append(c[1]) pl.figure() mpu.plot_yx(area_list,z_list,linewidth=2,label='area') pl.savefig('hist_area.png') pl.figure() mpu.plot_yx(n_points_list,z_list,linewidth=2,color='g',label='n_points') # pl.legend(loc='best') pl.xlabel('Z coordinate (m)') pl.ylabel('# points') pl.savefig('hist_points.png') ## find the x and y coord of the center of the circle of given radius that # best matches the data. # @param rad - radius of the circle (not optimized) # @param x_guess - guess for x coord of center # @param y_guess - guess for y coord of center. # @param pts - 2xN np matrix of points. # @return x,y (x and y coord of the center of the circle) def fit_rotary_joint(rad,x_guess,y_guess,pts): def error_function(params): center = np.matrix((params[0],params[1])).T #print 'pts.shape', pts.shape #print 'center.shape', center.shape #print 'ut.norm(pts-center).shape',ut.norm(pts-center).shape err = ut.norm(pts-center).A1 - rad res = np.dot(err,err) return res params_1 = [x_guess,y_guess] r = so.fmin_bfgs(error_function,params_1,full_output=1) opt_params_1,f_opt_1 = r[0],r[1] params_2 = [x_guess,y_guess+2*rad] r = so.fmin_bfgs(error_function,params_2,full_output=1) opt_params_2,f_opt_2 = r[0],r[1] if f_opt_2<f_opt_1: return opt_params_2[0],opt_params_2[1] else: return opt_params_1[0],opt_params_1[1] ## find the x and y coord of the center of the circle and the radius that # best matches the data. # @param rad_guess - guess for the radius of the circle # @param x_guess - guess for x coord of center # @param y_guess - guess for y coord of center. # @param pts - 2xN np matrix of points. # @param method - optimization method. ('fmin' or 'fmin_bfgs') # @param verbose - passed onto the scipy optimize functions. whether to print out the convergence info. # @return r,x,y (radius, x and y coord of the center of the circle) def fit_circle(rad_guess,x_guess,y_guess,pts,method,verbose=True): def error_function(params): center = np.matrix((params[0],params[1])).T rad = params[2] #print 'pts.shape', pts.shape #print 'center.shape', center.shape #print 'ut.norm(pts-center).shape',ut.norm(pts-center).shape err = ut.norm(pts-center).A1 - rad res = np.dot(err,err) return res params_1 = [x_guess,y_guess,rad_guess] if method == 'fmin': r = so.fmin(error_function,params_1,xtol=0.0002,ftol=0.000001,full_output=1,disp=verbose) opt_params_1,fopt_1 = r[0],r[1] elif method == 'fmin_bfgs': r = so.fmin_bfgs(error_function,params_1,full_output=1,disp=verbose) opt_params_1,fopt_1 = r[0],r[1] else: raise RuntimeError('unknown method: '+method) params_2 = [x_guess,y_guess+2*rad_guess,rad_guess] if method == 'fmin': r = so.fmin(error_function,params_2,xtol=0.0002,ftol=0.000001,full_output=1,disp=verbose) opt_params_2,fopt_2 = r[0],r[1] elif method == 'fmin_bfgs': r = so.fmin_bfgs(error_function,params_2,full_output=1,disp=verbose) opt_params_2,fopt_2 = r[0],r[1] else: raise RuntimeError('unknown method: '+method) if fopt_2<fopt_1: return opt_params_2[2],opt_params_2[0],opt_params_2[1] else: return opt_params_1[2],opt_params_1[0],opt_params_1[1] if __name__ == '__main__': p = optparse.OptionParser() p.add_option('-f', action='store', type='string', dest='fname', help='pkl file to use.', default='') p.add_option('--xy', action='store_true', dest='xy', help='plot the x and y coordinates of the end effector.') p.add_option('--yz', action='store_true', dest='yz', help='plot the y and z coordinates of the end effector.') p.add_option('--xz', action='store_true', dest='xz', help='plot the x and z coordinates of the end effector.') p.add_option('--plot_ellipses', action='store_true', dest='plot_ellipses', help='plot the stiffness ellipse in the x-y plane') p.add_option('--pfc', action='store_true', dest='pfc', help='plot the radial and tangential components of the force.') p.add_option('--pmf', action='store_true', dest='pmf', help='plot things with the mechanism alinged with the axes.') p.add_option('--pff', action='store_true', dest='pff', help='plot the force field corresponding to a stiffness ellipse.') p.add_option('--pev', action='store_true', dest='pev', help='plot the stiffness ellipses for different combinations of the rel stiffnesses.') p.add_option('--plot_forces', action='store_true', dest='plot_forces', help='plot the force in the x-y plane') p.add_option('--plot_forces_error', action='store_true', dest='plot_forces_error', help='plot the error between the computed and measured (ATI) forces in the x-y plane') p.add_option('--xyz', action='store_true', dest='xyz', help='plot in 3d the coordinates of the end effector.') p.add_option('-r', action='store', type='float', dest='rad', help='radius of the joint.', default=None) p.add_option('--rad_fix', action='store_true', dest='rad_fix', help='do not optimize for the radius.') p.add_option('--noshow', action='store_true', dest='noshow', help='do not display the figure (use while saving figures to disk)') p.add_option('--exptplot', action='store_true', dest='exptplot', help='put all the graphs of an experiment as subplots.') p.add_option('--pwf', action='store_true', dest='pwf', help='plot the workspace at some z coord.') opt, args = p.parse_args() fname = opt.fname xy_flag = opt.xy yz_flag = opt.yz xz_flag = opt.xz plot_forces_flag = opt.plot_forces plot_ellipses_flag = opt.plot_ellipses plot_forces_error_flag = opt.plot_forces_error plot_force_components_flag = opt.pfc plot_force_field_flag = opt.pff plot_mechanism_frame = opt.pmf xyz_flag = opt.xyz rad = opt.rad show_fig = not(opt.noshow) plot_ellipses_vary_flag = opt.pev expt_plot = opt.exptplot rad_fix = opt.rad_fix plot_workspace_flag = opt.pwf if plot_workspace_flag: compute_workspace_z() # hull = compute_workspace(z=-0.22,plot=True) # pl.show() if plot_ellipses_vary_flag: show_fig=False i = 0 ratio_list1 = [0.1,0.3,0.5,0.7,0.9] # coarse search ratio_list2 = [0.1,0.3,0.5,0.7,0.9] # coarse search ratio_list3 = [0.1,0.3,0.5,0.7,0.9] # coarse search # ratio_list1 = [0.7,0.8,0.9,1.0] # ratio_list2 = [0.7,0.8,0.9,1.0] # ratio_list3 = [0.3,0.4,0.5,0.6,0.7] # ratio_list1 = [1.0,2.,3.0] # ratio_list2 = [1.,2.,3.] # ratio_list3 = [0.3,0.4,0.5,0.6,0.7] inv_mean_list,std_list = [],[] x_l,y_l,z_l = [],[],[] s0 = 0.2 #s0 = 0.4 for s1 in ratio_list1: for s2 in ratio_list2: for s3 in ratio_list3: i += 1 s_list = [s0,s1,s2,s3,0.8] #s_list = [s1,s2,s3,s0,0.8] print '################## s_list:', s_list m,s = plot_stiff_ellipse_map(s_list,i) inv_mean_list.append(1./m) std_list.append(s) x_l.append(s1) y_l.append(s2) z_l.append(s3) ut.save_pickle({'x_l':x_l,'y_l':y_l,'z_l':z_l,'inv_mean_list':inv_mean_list,'std_list':std_list}, 'stiff_dict_'+ut.formatted_time()+'.pkl') d3m.plot_points(np.matrix([x_l,y_l,z_l]),scalar_list=inv_mean_list,mode='sphere') mlab.axes() d3m.show() sys.exit() if fname=='': print 'please specify a pkl file (-f option)' print 'Exiting...' sys.exit() d = ut.load_pickle(fname) actual_cartesian = joint_to_cartesian(d['actual']) eq_cartesian = joint_to_cartesian(d['eq_pt']) for p in actual_cartesian.p_list: print p[0],p[1],p[2] if rad != None: #rad = 0.39 # lab cabinet recessed. #rad = 0.42 # kitchen cabinet #rad = 0.80 # lab glass door pts_list = actual_cartesian.p_list ee_start_pos = pts_list[0] x_guess = ee_start_pos[0] y_guess = ee_start_pos[1] - rad print 'before call to fit_rotary_joint' pts_2d = (np.matrix(pts_list).T)[0:2,:] t0 = time.time() if rad_fix: rad_guess = 0.9 else: rad_guess = rad rad_fmin,cx,cy = fit_circle(rad_guess,x_guess,y_guess,pts_2d[:,0:-4],method='fmin') t1 = time.time() rad_opt,cx,cy = fit_circle(rad_guess,x_guess,y_guess,pts_2d[:,0:-4],method='fmin_bfgs') t2 = time.time() print 'after fit_rotary_joint' print 'optimized radius:', rad_opt print 'optimized radius fmin:', rad_fmin print 'time to bfgs:', t2-t1 print 'time to fmin:', t1-t0 if rad_fix: cx,cy = fit_rotary_joint(rad,x_guess,y_guess,pts_2d[:,0:-4]) else: rad = rad_opt if plot_mechanism_frame: if expt_plot: pl.subplot(231) # transform points so that the mechanism is in a fixed position. start_pt = actual_cartesian.p_list[0] x_diff = start_pt[0] - cx y_diff = start_pt[1] - cy angle = math.atan2(y_diff,x_diff) - math.radians(90) rot_mat = tr.Rz(angle)[0:2,0:2] translation_mat = np.matrix([cx,cy]).T robot_width,robot_length = 0.1,0.2 robot_x_list = [-robot_width/2,-robot_width/2,robot_width/2,robot_width/2,-robot_width/2] robot_y_list = [-robot_length/2,robot_length/2,robot_length/2,-robot_length/2,-robot_length/2] robot_mat = rot_mat*(np.matrix([robot_x_list,robot_y_list]) - translation_mat) mpu.plot_yx(robot_mat[1,:].A1,robot_mat[0,:].A1,linewidth=2,scatter_size=0, color='k',label='torso') pts2d_actual = (np.matrix(actual_cartesian.p_list).T)[0:2] pts2d_actual_t = rot_mat *(pts2d_actual - translation_mat) mpu.plot_yx(pts2d_actual_t[1,:].A1,pts2d_actual_t[0,:].A1,scatter_size=20,label='FK') end_pt = pts2d_actual_t[:,-1] end_angle = tr.angle_within_mod180(math.atan2(end_pt[1,0],end_pt[0,0])-math.radians(90)) mpu.plot_circle(0,0,rad,0.,end_angle,label='Actual_opt',color='r') mpu.plot_radii(0,0,rad,0.,end_angle,interval=math.radians(15),color='r') pl.legend(loc='best') pl.axis('equal') if not(expt_plot): str_parts = fname.split('.') fig_name = str_parts[0]+'_robot_pose.png' pl.savefig(fig_name) pl.figure() else: pl.subplot(232) pl.text(0.1,0.15,d['info']) pl.text(0.1,0.10,'control: '+d['strategy']) pl.text(0.1,0.05,'robot angle: %.2f'%math.degrees(angle)) pl.text(0.1,0,'optimized radius: %.2f'%rad_opt) pl.text(0.1,-0.05,'radius used: %.2f'%rad) pl.text(0.1,-0.10,'opening angle: %.2f'%math.degrees(end_angle)) s_list = d['stiffness'].stiffness_list s_scale = d['stiffness'].stiffness_scale sl = [min(s*s_scale,1.0) for s in s_list] pl.text(0.1,-0.15,'stiffness list: %.2f, %.2f, %.2f, %.2f'%(sl[0],sl[1],sl[2],sl[3])) pl.text(0.1,-0.20,'stop condition: '+d['result']) time_dict = d['time_dict'] pl.text(0.1,-0.25,'time to hook: %.2f'%(time_dict['before_hook']-time_dict['before_pull'])) pl.text(0.1,-0.30,'time to pull: %.2f'%(time_dict['before_pull']-time_dict['after_pull'])) pl.ylim(-0.45,0.25) if not(expt_plot): pl.figure() if xy_flag: st_pt = pts_2d[:,0] start_angle = tr.angle_within_mod180(math.atan2(st_pt[1,0]-cy,st_pt[0,0]-cx) - math.radians(90)) end_pt = pts_2d[:,-1] end_angle = tr.angle_within_mod180(math.atan2(end_pt[1,0]-cy,end_pt[0,0]-cx) - math.radians(90)) print 'start_angle, end_angle:', math.degrees(start_angle), math.degrees(end_angle) print 'angle through which mechanism turned:', math.degrees(end_angle-start_angle) if expt_plot: pl.subplot(233) plot_cartesian(actual_cartesian,xaxis=0,yaxis=1,color='b',label='End Effector Trajectory') plot_cartesian(eq_cartesian, xaxis=0,yaxis=1,color='g',label='Eq Point') mpu.plot_circle(cx,cy,rad,start_angle,end_angle,label='Estimated Kinematics',color='r') # if rad<0.6: # mpu.plot_radii(cx,cy,rad,start_angle,end_angle,interval=math.radians(100),color='r') # pl.title(d['info']) leg = pl.legend(loc='best')#,handletextsep=0.020,handlelen=0.003,labelspacing=0.003) leg.draw_frame(False) ax = pl.gca() ax.set_xlim(ax.get_xlim()[::-1]) ax.set_ylim(ax.get_ylim()[::-1]) force_traj = d['force'] forces = np.matrix(force_traj.f_list).T force_mag = ut.norm(forces) print 'force_mag:', force_mag.A1 elif yz_flag: plot_cartesian(actual_cartesian,xaxis=1,yaxis=2,color='b',label='FK') plot_cartesian(eq_cartesian, xaxis=1,yaxis=2,color='g',label='Eq Point') elif xz_flag: plot_cartesian(actual_cartesian,xaxis=0,yaxis=2,color='b',label='FK') plot_cartesian(eq_cartesian, xaxis=0,yaxis=2,color='g',label='Eq Point') if plot_forces_flag or plot_forces_error_flag or plot_ellipses_flag or plot_force_components_flag or plot_force_field_flag: arm_stiffness_list = d['stiffness'].stiffness_list scale = d['stiffness'].stiffness_scale asl = [min(scale*s,1.0) for s in arm_stiffness_list] ftraj_jinv,ftraj_stiff,ftraj_torque,k_cart_list=compute_forces(d['actual'],d['eq_pt'], d['torque'],asl) if plot_forces_flag: plot_forces_quiver(actual_cartesian,d['force'],color='k') plot_forces_quiver(actual_cartesian,ftraj_jinv,color='y') #plot_forces_quiver(actual_cartesian,ftraj_stiff,color='y') if plot_ellipses_flag: #plot_stiff_ellipses(k_cart_list,actual_cartesian) if expt_plot: subplotnum=234 else: pl.figure() subplotnum=111 plot_stiff_ellipses(k_cart_list,eq_cartesian,subplotnum=subplotnum) if plot_forces_error_flag: plot_error_forces(d['force'].f_list,ftraj_jinv.f_list) #plot_error_forces(d['force'].f_list,ftraj_stiff.f_list) if plot_force_components_flag: p_list = actual_cartesian.p_list frad_list,ftan_list = compute_radial_tangential_forces(d['force'].f_list,p_list,cx,cy) if expt_plot: pl.subplot(235) else: pl.figure() time_list = d['force'].time_list time_list = [t-time_list[0] for t in time_list] x_coord_list = np.matrix(p_list)[:,0].A1.tolist() mpu.plot_yx(frad_list,x_coord_list,scatter_size=50,color=time_list,cb_label='time') pl.xlabel('x coord of end effector (m)') pl.ylabel('magnitude of radial force (N)') pl.title(d['info']) if expt_plot: pl.subplot(236) else: pl.figure() mpu.plot_yx(ftan_list,x_coord_list,scatter_size=50,color=time_list,cb_label='time') pl.xlabel('x coord of end effector (m)') pl.ylabel('magnitude of tangential force (N)') pl.title(d['info']) if plot_force_field_flag: plot_stiffness_field(k_cart_list[0],plottitle='start') plot_stiffness_field(k_cart_list[-1],plottitle='end') if expt_plot: f = pl.gcf() curr_size = f.get_size_inches() f.set_size_inches(curr_size[0]*2,curr_size[1]*2) str_parts = fname.split('.') if d.has_key('strategy'): fig_name = str_parts[0]+'_'+d['strategy']+'.png' else: fig_name = str_parts[0]+'_res.png' f.savefig(fig_name) if show_fig: pl.show() else: print '################################' print 'show_fig is FALSE' if xyz_flag: plot_cartesian(traj, xaxis=0,yaxis=1,zaxis=2) mlab.show()
[ [ 1, 0, 0.0332, 0.0011, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0332, 0.0011, 0, 0.66, 0.0345, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0354, 0.0011, 0, 0....
[ "import roslib; roslib.load_manifest('2009_humanoids_epc_pull')", "import roslib; roslib.load_manifest('2009_humanoids_epc_pull')", "import scipy.optimize as so", "import math, numpy as np", "import pylab as pl", "import sys, optparse, time", "import copy", "from enthought.mayavi import mlab", "impo...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # Author: Advait Jain import m3.toolbox as m3t import hokuyo.hokuyo_scan as hs import tilting_hokuyo.tilt_hokuyo_servo as ths import mekabot.hrl_robot as hr import hrl_lib.util as ut, hrl_lib.transforms as tr import util as uto import camera import sys, time, os, optparse import math, numpy as np import copy import arm_trajectories as at from opencv.cv import * from opencv.highgui import * from threading import RLock import threading hook_3dprint_angle = math.radians(20-2.54) class CompliantMotionGenerator(threading.Thread): ''' a specific form of compliant motion. class name might be inappropriate. ''' def __init__(self): # stiffness in Nm/rad: [20,50,15,25] self.settings_r = hr.MekaArmSettings(stiffness_list=[0.1939,0.6713,0.748,0.7272,0.8]) # self.settings_r = hr.MekaArmSettings(stiffness_list=[0.2,1.0,1.0,0.4,0.8]) self.settings_stiff = hr.MekaArmSettings(stiffness_list=[0.8,1.0,1.0,1.0,0.8]) self.firenze = hr.M3HrlRobot(connect=True,right_arm_settings=self.settings_stiff) self.hok = hs.Hokuyo('utm',0,flip=True) self.thok = ths.tilt_hokuyo('/dev/robot/servo0',5,self.hok,l1=0.,l2=-0.055) self.cam = camera.Camera('mekabotUTM') self.set_camera_settings() self.fit_circle_lock = RLock() threading.Thread.__init__(self) def run(self): self.circle_estimator() def stop(self): self.run_fit_circle_thread = False self.firenze.stop() ## pose the arm by moving the end effector to the hookable location. # @param hook_angle - RADIANS(0, -90, 90 etc.) # 0 - horizontal, -pi/2 hook points up, +pi/2 hook points down def pose_arm(self,hook_angle): print 'press ENTER to pose the robot.' k=m3t.get_keystroke() if k!='\r': print 'You did not press ENTER.' return # settings_r_orig = copy.copy(self.firenze.arm_settings['right_arm']) settings_torque_gc = hr.MekaArmSettings(stiffness_list=[0.,0.,0.,0.,0.],control_mode='torque_gc') self.firenze.set_arm_settings(settings_torque_gc,None) self.firenze.step() print 'hit ENTER to end posing, something else to exit' k=m3t.get_keystroke() p = self.firenze.end_effector_pos('right_arm') q = self.firenze.get_joint_angles('right_arm') # self.firenze.set_arm_settings(settings_r_orig,None) self.firenze.set_arm_settings(self.settings_stiff,None) self.firenze.set_joint_angles('right_arm',q) self.firenze.step() self.firenze.set_joint_angles('right_arm',q) self.firenze.step() rot_mat = tr.Rz(hook_angle-hook_3dprint_angle)*tr.Ry(math.radians(-90)) self.firenze.go_cartesian('right_arm',p,rot_mat,speed=0.1) print 'hit ENTER after making finer adjustment, something else to exit' k=m3t.get_keystroke() p = self.firenze.end_effector_pos('right_arm') q = self.firenze.get_joint_angles('right_arm') self.firenze.set_joint_angles('right_arm',q) self.firenze.step() def set_camera_settings(self): self.cam.set_frame_rate(30) self.cam.set_auto() self.cam.set_gamma(1) self.cam.set_whitebalance(r_val=512,b_val=544) def compliant_motion(self,equi_pt_generator,time_step,rapid_call_func=None): ''' equi_pt_generator: function that returns stop,q q: list of 7 joint angles stop: string which is '' for compliant motion to continue rapid_call_func: called in the time between calls to the equi_pt_generator can be used for logging, safety etc. returns stop stop: string which is '' for compliant motion to continue time_step: time between successive calls to equi_pt_generator returns stop (the string which has the reason why the compliant motion stopped.) ''' stop,q = equi_pt_generator() while stop == '': self.firenze.set_joint_angles('right_arm',q) t1 = time.time() t_end = t1+time_step while t1<t_end: self.firenze.step() if rapid_call_func != None: stop = rapid_call_func() if stop != '': break t1 = time.time() if stop != '': break stop,q = equi_pt_generator() return stop ## log the joint angles, equi pt joint angles and forces. def log_state(self): t_now = time.time() q_now = self.firenze.get_joint_angles('right_arm') qdot_now = self.firenze.get_joint_velocities('right_arm') tau_now = self.firenze.get_joint_torques('right_arm') self.jt_torque_trajectory.q_list.append(tau_now) self.jt_torque_trajectory.time_list.append(t_now) self.pull_trajectory.q_list.append(q_now) self.pull_trajectory.qdot_list.append(qdot_now) self.pull_trajectory.time_list.append(t_now) #self.eq_pt_trajectory.p_list.append(self.eq_pt_cartesian.A1.tolist()) self.eq_pt_trajectory.q_list.append(self.q_guess) # see equi_pt_generator - q_guess is the config for the eq point. self.eq_pt_trajectory.time_list.append(t_now) wrist_force = self.firenze.get_wrist_force('right_arm',base_frame=True) self.force_trajectory.f_list.append(wrist_force.A1.tolist()) self.force_trajectory.time_list.append(t_now) def common_stopping_conditions(self): stop = '' if self.q_guess == None: stop = 'IK fail' wrist_force = self.firenze.get_wrist_force('right_arm',base_frame=True) mag = np.linalg.norm(wrist_force) if mag > self.eq_force_threshold: stop = 'force exceed' if mag < 1.2 and self.hooked_location_moved: if (self.prev_force_mag - mag) > 10.: stop = 'slip: force step decrease and below thresold.' else: self.slip_count += 1 else: self.slip_count = 0 if self.slip_count == 4: stop = 'slip: force below threshold for too long.' curr_pos = self.firenze.FK('right_arm',self.pull_trajectory.q_list[-1]) if curr_pos[0,0]<0.27 and curr_pos[1,0]>-0.2: stop = 'danger of self collision' return stop def update_eq_point(self,motion_vec,step_size): next_pt = self.eq_pt_cartesian + step_size * motion_vec rot_mat = self.eq_IK_rot_mat # self.q_guess[1] += math.radians(1) q_eq = self.firenze.IK('right_arm',next_pt,rot_mat,self.q_guess) self.eq_pt_cartesian = next_pt self.q_guess = q_eq return q_eq def circle_estimator(self): self.run_fit_circle_thread = True print 'Starting the circle estimating thread.' while self.run_fit_circle_thread: self.fit_circle_lock.acquire() if len(self.cartesian_pts_list)==0: self.fit_circle_lock.release() continue pts_2d = (np.matrix(self.cartesian_pts_list).T)[0:2,:] self.fit_circle_lock.release() rad = self.rad_guess start_pos = self.firenze.FK('right_arm',self.pull_trajectory.q_list[0]) rad,cx,cy = at.fit_circle(rad,start_pos[0,0],start_pos[1,0]-rad,pts_2d,method='fmin_bfgs',verbose=False) rad = ut.bound(rad,3.0,0.1) self.fit_circle_lock.acquire() self.cx = cx self.cy = cy # self.rad = rad self.fit_circle_lock.release() print 'Ended the circle estimating thread.' ## constantly update the estimate of the kinematics and move the # equilibrium point along the tangent of the estimated arc, and # try to keep the radial force constant. def equi_pt_generator_control_radial_force(self): self.log_state() q_eq = self.update_eq_point(self.eq_motion_vec,0.01) stop = self.common_stopping_conditions() wrist_force = self.firenze.get_wrist_force('right_arm',base_frame=True) mag = np.linalg.norm(wrist_force) start_pos = self.firenze.FK('right_arm',self.pull_trajectory.q_list[0]) curr_pos = self.firenze.FK('right_arm',self.pull_trajectory.q_list[-1]) if (start_pos[0,0]-curr_pos[0,0])>0.09 and self.hooked_location_moved==False: # change the force threshold once the hook has started pulling. self.hooked_location_moved = True self.eq_force_threshold = ut.bound(mag+30.,20.,80.) self.piecewise_force_threshold = ut.bound(mag+5.,0.,80.) self.fit_circle_lock.acquire() self.cartesian_pts_list.append(curr_pos.A1.tolist()) self.fit_circle_lock.release() # find tangential direction. radial_vec = curr_pos[0:2]-np.matrix([self.cx,self.cy]).T radial_vec = radial_vec/np.linalg.norm(radial_vec) tan_x,tan_y = -radial_vec[1,0],radial_vec[0,0] if tan_x >0.: tan_x = -tan_x tan_y = -tan_y self.eq_motion_vec = np.matrix([tan_x,tan_y,0.]).T f_vec = -1*np.array([wrist_force[0,0],wrist_force[1,0]]) f_rad_mag = np.dot(f_vec,radial_vec.A1) #if f_rad_mag>10.: if f_rad_mag>5.: self.eq_motion_vec[0:2] -= radial_vec/2. * self.hook_maintain_dist_plane/0.05 else: self.eq_motion_vec[0:2] += radial_vec/2. * self.hook_maintain_dist_plane/0.05 self.prev_force_mag = mag return stop,q_eq ## moves eq point along the -x axis. def equi_pt_generator_line(self): self.log_state() #q_eq = self.update_eq_point(self.eq_motion_vec,0.005) q_eq = self.update_eq_point(self.eq_motion_vec,0.010) stop = self.common_stopping_conditions() wrist_force = self.firenze.get_wrist_force('right_arm',base_frame=True) mag = np.linalg.norm(wrist_force) start_pos = self.firenze.FK('right_arm',self.pull_trajectory.q_list[0]) curr_pos = self.firenze.FK('right_arm',self.pull_trajectory.q_list[-1]) if (start_pos[0,0]-curr_pos[0,0])>0.09 and self.hooked_location_moved==False: # change the force threshold once the hook has started pulling. self.hooked_location_moved = True self.eq_force_threshold = ut.bound(mag+15.,20.,80.) self.prev_force_mag = mag return stop,q_eq ## move the end effector to properly hook onto the world # direction of motion governed by the hook angle. # @param hook_angle - angle of hook in RADIANS (see pose_arm or pull for details.) def get_firm_hook(self, hook_angle): rot_mat = tr.Rz(hook_angle-hook_3dprint_angle)*tr.Ry(math.radians(-90)) # move in the +x until contact. vec = np.matrix([0.08,0.,0.]).T self.firenze.move_till_hit('right_arm',vec=vec,force_threshold=2.0,rot=rot_mat, speed=0.05) # now move in direction of hook. vec = tr.rotX(-hook_angle) * np.matrix([0.,0.05,0.]).T self.firenze.move_till_hit('right_arm',vec=vec,force_threshold=5.0,rot=rot_mat, speed=0.05,bias_FT=False) self.firenze.set_arm_settings(self.settings_r,None) self.firenze.step() def pull(self,hook_angle,force_threshold,use_utm=False,use_camera=False,strategy='line_neg_x', pull_loc=None, info_string=''): ''' force_threshold - max force at which to stop pulling. hook_angle - radians(0, -90, 90 etc.) 0 - horizontal, -pi/2 hook points up, +pi/2 hook points down use_utm - to take 3D scans or not. use_camera - to take pictures from the camera or not. strategy - 'line_neg_x': move eq point along -x axis. 'piecewise_linear': try and estimate circle and move along it. 'control_radial_force': try and keep the radial force constant 'control_radial_dist' pull_loc - 3x1 np matrix of location for pulling. If None then arm will go into gravity comp and user can show the location. info_string - string saved with key 'info' in the pkl. ''' if use_utm: self.firenze.step() armconfig1 = self.firenze.get_joint_angles('right_arm') plist1,slist1 = self.scan_3d() if use_camera: cam_plist1, cam_imlist1 = self.image_region() else: cam_plist1,cam_imlist1 = None,None rot_mat = tr.Rz(hook_angle-hook_3dprint_angle)*tr.Ry(math.radians(-90)) if pull_loc == None: self.pose_arm(hook_angle) pull_loc = self.firenze.end_effector_pos('right_arm') ut.save_pickle(pull_loc,'pull_loc_'+info_string+'_'+ut.formatted_time()+'.pkl') else: pt1 = copy.copy(pull_loc) pt1[0,0] = pt1[0,0]-0.1 print 'pt1:', pt1.A1 print 'pull_loc:', pull_loc.A1 self.firenze.go_cartesian('right_arm',pt1,rot_mat,speed=0.2) self.firenze.go_cartesian('right_arm',pull_loc,rot_mat,speed=0.07) print 'press ENTER to pull' k=m3t.get_keystroke() if k != '\r': return time_dict = {} time_dict['before_hook'] = time.time() print 'first getting a good hook' self.get_firm_hook(hook_angle) time.sleep(0.5) time_dict['before_pull'] = time.time() print 'pull begins' stiffness_scale = self.settings_r.stiffness_scale vec = tr.rotX(-hook_angle) * np.matrix([0.,0.05/stiffness_scale,0.]).T self.keep_hook_vec = vec self.hook_maintain_dist_plane = np.dot(vec.A1,np.array([0.,1.,0.])) self.eq_pt_cartesian = self.firenze.end_effector_pos('right_arm') + vec q_eq = self.firenze.IK('right_arm',self.eq_pt_cartesian,rot_mat) self.firenze.go_jointangles('right_arm',q_eq,speed=math.radians(30)) self.q_guess = q_eq # self.q_guess = self.firenze.get_joint_angles('right_arm') self.pull_trajectory = at.JointTrajectory() self.jt_torque_trajectory = at.JointTrajectory() self.eq_pt_trajectory = at.JointTrajectory() self.force_trajectory = at.ForceTrajectory() self.firenze.step() start_config = self.firenze.get_joint_angles('right_arm') self.eq_IK_rot_mat = rot_mat # for equi pt generators. self.eq_force_threshold = force_threshold self.hooked_location_moved = False # flag to indicate when the hooking location started moving. self.prev_force_mag = np.linalg.norm(self.firenze.get_wrist_force('right_arm')) self.eq_motion_vec = np.matrix([-1.,0.,0.]).T self.slip_count = 0 if strategy == 'line_neg_x': result = self.compliant_motion(self.equi_pt_generator_line,0.025) elif strategy == 'control_radial_force': self.cartesian_pts_list = [] self.piecewise_force_threshold = force_threshold self.rad_guess = 1.0 self.cx = 0.6 self.cy = -self.rad_guess self.start() # start the circle estimation thread result = self.compliant_motion(self.equi_pt_generator_control_radial_force,0.025) else: raise RuntimeError('unknown pull strategy: ', strategy) if result == 'slip: force step decrease' or result == 'danger of self collision': self.firenze.motors_off() print 'powered off the motors.' print 'Compliant motion result:', result print 'Original force threshold:', force_threshold print 'Adapted force threshold:', self.eq_force_threshold time_dict['after_pull'] = time.time() d = {'actual': self.pull_trajectory, 'eq_pt': self.eq_pt_trajectory, 'force': self.force_trajectory, 'torque': self.jt_torque_trajectory, 'stiffness': self.firenze.arm_settings['right_arm'], 'info': info_string, 'force_threshold': force_threshold, 'eq_force_threshold': self.eq_force_threshold, 'hook_angle':hook_angle, 'result':result,'strategy':strategy,'time_dict':time_dict} self.firenze.step() armconfig2 = self.firenze.get_joint_angles('right_arm') if use_utm: plist2,slist2 = self.scan_3d() d['start_config']=start_config d['armconfig1']=armconfig1 d['armconfig2']=armconfig2 d['l1'],d['l2']=0.,-0.055 d['scanlist1'],d['poslist1']=slist1,plist1 d['scanlist2'],d['poslist2']=slist2,plist2 d['cam_plist1']=cam_plist1 d['cam_imlist1']=cam_imlist1 ut.save_pickle(d,'pull_trajectories_'+d['info']+'_'+ut.formatted_time()+'.pkl') def scan_3d(self): tilt_angles = (math.radians(20),math.radians(70)) pos_list,scan_list = self.thok.scan(tilt_angles,speed=math.radians(10),save_scan=False) return pos_list,scan_list def save_frame(self): cvim = self.cam.get_frame() cvSaveImage('im_'+ut.formatted_time()+'.png',cvim) def image_region(self): ''' takes images from the UTM camera at different angles. returns list of servo angles, list of images. images are numpy images. so that they can be pickled. ''' im_list = [] p_list = [] for cmd_ang in [0,30,45]: self.thok.servo.move_angle(math.radians(cmd_ang)) cvim = self.cam.get_frame() cvim = self.cam.get_frame() im_list.append(uto.cv2np(cvim,format='BGR')) p_list.append(self.thok.servo.read_angle()) self.thok.servo.move_angle(math.radians(0)) return p_list,im_list def test_IK(rot_mat): ''' try out the IK at a number of different cartesian points in the workspace, with the given rotation matrix for the end effector. ''' print 'press ENTER to start.' k=m3t.get_keystroke() while k=='\r': p = firenze.end_effector_pos('right_arm') firenze.go_cartesian('right_arm',p,rot_mat,speed=0.1) firenze.step() print 'press ENTER to save joint angles.' k=m3t.get_keystroke() if k == '\r': firenze.step() q = firenze.get_joint_angles('right_arm') ut.save_pickle(q,'arm_config_'+ut.formatted_time()+'.pkl') print 'press ENTER for next IK test. something else to exit.' k=m3t.get_keystroke() def test_elbow_angle(): firenze = hr.M3HrlRobot(connect=False) hook_3dprint_angle = math.radians(20-2.54) rot_mat = tr.Rz(math.radians(-90.)-hook_3dprint_angle)*tr.Ry(math.radians(-90)) x_list = [0.55,0.45,0.35] y = -0.2 z = -0.23 for x in x_list: p0 = np.matrix([x,y,z]).T q = firenze.IK('right_arm',p0,rot_mat) # q[1] = math.radians(15) # q = firenze.IK('right_arm',p0,rot_mat,q_guess = q) elbow_angle = firenze.elbow_plane_angle('right_arm',q) print '---------------------------------------' print 'ee position:', p0.A1 # print 'joint angles:', q print 'elbow_angle:', math.degrees(elbow_angle) if __name__=='__main__': p = optparse.OptionParser() p.add_option('--ik_single_pos', action='store_true', dest='ik_single_pos', help='test IK at a single position.') p.add_option('--ik_test', action='store_true', dest='ik_test', help='test IK in a loop.') p.add_option('--pull', action='store_true', dest='pull', help='pull with hook up (name will be changed later).') p.add_option('--pull_pos', action='store', type='string', dest='pull_pos_pkl', help='pkl file with 3D coord of point to start pulling at.', default='') p.add_option('--firm_hook', action='store_true', dest='firm_hook', help='test getting a firm hook on things.') p.add_option('--scan', action='store_true', dest='scan', help='take and save 3D scans. specify --pull also.') p.add_option('--camera', action='store_true', dest='camera', help='take and save images from UTM camera. specify --pull also.') p.add_option('--ha', action='store', dest='ha',type='float', default=None,help='hook angle (degrees).') p.add_option('--ft', action='store', dest='ft',type='float', default=None,help='force threshold (Newtons).') p.add_option('--info', action='store', type='string', dest='info_string', help='string to save in the pkl log.', default='') p.add_option('--ve', action='store_true', dest='ve', help='vary experiment. (vary stiffness settings and repeatedly pull)') p.add_option('--eaf', action='store_true', dest='eaf', help='test elbow angle finding with the horizontal plane.') opt, args = p.parse_args() ik_single_pos_flag = opt.ik_single_pos test_ik_flag = opt.ik_test pull_flag = opt.pull pull_pos_pkl = opt.pull_pos_pkl firm_hook_flag = opt.firm_hook scan_flag = opt.scan camera_flag = opt.camera ha = opt.ha ft = opt.ft info_string = opt.info_string vary_expt_flag = opt.ve elbow_angle_flag = opt.eaf try: if vary_expt_flag: stiff_scale_list = [1.0,1.2,0.8] if pull_pos_pkl != '': pull_loc = ut.load_pickle(pull_pos_pkl) else: raise RuntimeError('Need to specify a pull_pos with vary_expt') cmg = CompliantMotionGenerator() print 'hit a key to power up the arms.' k=m3t.get_keystroke() cmg.firenze.power_on() #for strategy in ['line_neg_x','control_radial_force']: for strategy in ['line_neg_x','control_radial_dist','control_radial_force']: #for strategy in ['line_neg_x']: #for strategy in ['piecewise_linear']: for s_scale in stiff_scale_list: cmg.settings_r.stiffness_scale = s_scale cmg.pull(math.radians(ha), ft,use_utm=scan_flag,use_camera=camera_flag, strategy=strategy,pull_loc=pull_loc,info_string=info_string) cmg.firenze.maintain_configuration() cmg.firenze.motors_on() cmg.firenze.set_arm_settings(cmg.settings_stiff,None) time.sleep(0.5) print 'hit a key to end everything' k=m3t.get_keystroke() cmg.firenze.stop() sys.exit() if pull_flag or firm_hook_flag: if ha == None: print 'please specify hook angle (--ha)' print 'Exiting...' sys.exit() if ft == None and pull_flag: print 'please specify force threshold (--ft) along with --pull' print 'Exiting...' sys.exit() cmg = CompliantMotionGenerator() print 'hit a key to power up the arms.' k=m3t.get_keystroke() cmg.firenze.power_on() if pull_flag: if pull_pos_pkl != '': pull_loc = ut.load_pickle(pull_pos_pkl) else: pull_loc = None # cmg.pull(math.radians(ha), ft,use_utm=scan_flag,use_camera=camera_flag, # strategy = 'control_radial_dist',pull_loc=pull_loc,info_string=info_string) # cmg.pull(math.radians(ha), ft,use_utm=scan_flag,use_camera=camera_flag, # strategy = 'piecewise_linear',pull_loc=pull_loc,info_string=info_string) cmg.pull(math.radians(ha), ft,use_utm=scan_flag,use_camera=camera_flag, strategy = 'control_radial_force',pull_loc=pull_loc,info_string=info_string) # cmg.pull(math.radians(ha), ft,use_utm=scan_flag,use_camera=camera_flag, # strategy = 'line_neg_x',pull_loc=pull_loc,info_string=info_string) if firm_hook_flag: hook_angle = math.radians(ha) p = np.matrix([0.3,-0.25,-0.2]).T rot_mat = tr.Rz(hook_angle-hook_3dprint_angle)*tr.Ry(math.radians(-90)) cmg.firenze.go_cartesian('right_arm',p,rot_mat,speed=0.1) print 'hit a key to get a firm hook.' k=m3t.get_keystroke() cmg.get_firm_hook(hook_angle) print 'hit a key to end everything' k=m3t.get_keystroke() cmg.stop() # cmg = CompliantMotionGenerator() # print 'hit a key to test IK' # k=m3t.get_keystroke() # cmg.get_firm_hook(ha) #----------- non-class functions test -------------------- if elbow_angle_flag: test_elbow_angle() if ik_single_pos_flag or test_ik_flag: if ha == None: raise RuntimeError('You need to specify a hooking angle (--ha)') settings_r = hr.MekaArmSettings(stiffness_list=[0.15,0.7,0.8,0.8,0.8]) firenze = hr.M3HrlRobot(connect=True,right_arm_settings=settings_r) print 'hit a key to power up the arms.' k=m3t.get_keystroke() firenze.power_on() print 'hit a key to test IK' k=m3t.get_keystroke() #p = np.matrix([0.26,-0.25,-0.25]).T p = np.matrix([0.45,-0.2,-0.23]).T rot_mat = tr.Rz(math.radians(ha)-hook_3dprint_angle)*tr.Ry(math.radians(-90)) #rot_mat = tr.Rz(math.radians(0))*tr.Ry(math.radians(-90)) firenze.go_cartesian('right_arm',p,rot_mat,speed=0.1) if test_ik_flag: rot_mat = tr.Rz(math.radians(-110))*tr.Ry(math.radians(-90)) #rot_mat = tr.Rz(math.radians(0))*tr.Ry(math.radians(-90)) test_IK(rot_mat) print 'hit a key to end everything' k=m3t.get_keystroke() firenze.stop() except (KeyboardInterrupt, SystemExit): cmg.stop()
[ [ 1, 0, 0.0444, 0.0015, 0, 0.66, 0, 478, 0, 1, 0, 0, 478, 0, 0 ], [ 1, 0, 0.0473, 0.0015, 0, 0.66, 0.0526, 464, 0, 1, 0, 0, 464, 0, 0 ], [ 1, 0, 0.0488, 0.0015, 0, ...
[ "import m3.toolbox as m3t", "import hokuyo.hokuyo_scan as hs", "import tilting_hokuyo.tilt_hokuyo_servo as ths", "import mekabot.hrl_robot as hr", "import hrl_lib.util as ut, hrl_lib.transforms as tr", "import util as uto", "import camera", "import sys, time, os, optparse", "import math, numpy as np...
import roslib roslib.load_manifest("phantom_omni") import rospy import tf import math from geometry_msgs.msg import PoseStamped import pdb rospy.init_node("long_tip_pose_publisher") pub = rospy.Publisher('pr2_right', PoseStamped) r = rospy.Rate(60) while not rospy.is_shutdown(): ps = PoseStamped() ps.header.frame_id = 'omni1_link6' ps.header.stamp = rospy.get_rostime() ps.pose.position.x = -.134 ps.pose.position.y = 0 ps.pose.position.z = 0 q = tf.transformations.quaternion_from_euler(0, math.pi, 0) ps.pose.orientation.x = q[0] ps.pose.orientation.y = q[1] ps.pose.orientation.z = q[2] ps.pose.orientation.w = q[3] pub.publish(ps) r.sleep()
[ [ 1, 0, 0.0333, 0.0333, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0667, 0.0333, 0, 0.66, 0.1, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.1, 0.0333, 0, 0.66, ...
[ "import roslib", "roslib.load_manifest(\"phantom_omni\")", "import rospy", "import tf", "import math", "from geometry_msgs.msg import PoseStamped", "import pdb", "rospy.init_node(\"long_tip_pose_publisher\")", "pub = rospy.Publisher('pr2_right', PoseStamped)", "r = rospy.Rate(60)", "while not ro...
import roslib roslib.load_manifest("phantom_omni") import rospy import tf import tf.transformations as tr from geometry_msgs.msg import PoseStamped from geometry_msgs.msg import Pose from geometry_msgs.msg import Wrench import math import numpy as np def tf2mat(tf_trans): (trans, rot) = tf_trans return np.matrix(tr.translation_matrix(trans)) * np.matrix(tr.quaternion_matrix(rot)) def mat2pose(m): trans = tr.translation_from_matrix(m) rot = tr.quaternion_from_matrix(m) p = Pose() p.position.x = trans[0] p.position.y = trans[1] p.position.z = trans[2] p.orientation.x = rot[0] p.orientation.y = rot[1] p.orientation.z = rot[2] p.orientation.w = rot[3] return p rospy.init_node("omni_potential_well") wpub = rospy.Publisher('force_feedback', Wrench) r = rospy.Rate(100) listener = tf.TransformListener() listener.waitForTransform("/world", "/omni1_link6", rospy.Time(), rospy.Duration(4.0)) listener.waitForTransform("/world", "/sensable", rospy.Time(), rospy.Duration(4.0)) print 'running.' well_center = None angle = 0.; while not rospy.is_shutdown(): w_T_6 = tf2mat(listener.lookupTransform('/world', '/omni1_link6', rospy.Time(0))) qm = np.matrix(tr.quaternion_matrix(tr.quaternion_from_euler(0, math.pi, 0))) tm = np.matrix(tr.translation_matrix([-.134, 0, 0])) tip_6 = tm * qm tip_world = w_T_6 * tip_6 pos = np.matrix(tr.translation_from_matrix(tip_world)).T force_world = (w_T_6[0:3,0:3] * np.matrix([-2., 0, 0]).T) trans, rot = listener.lookupTransform('/sensable', '/world', rospy.Time(0)) quat_mat = np.matrix(tr.quaternion_matrix(rot)) force_sensable = quat_mat[0:3, 0:3] * force_world wr = Wrench() angle = np.radians(.3) + angle wr.force.x = force_sensable[0] wr.force.y = force_sensable[1] wr.force.z = force_sensable[2] print wr.force.z wpub.publish(wr) r.sleep() #ps = PoseStamped() #ps.header.frame_id = 'sensable' #ps.header.stamp = rospy.get_rostime() #ps.pose = mat2pose(tip_world) #pub.publish(ps) #print force_sensable.T #if well_center == None: # well_center = pos #else: # trans, rot = listener.lookupTransform('/world', '/sensable', rospy.Time(0)) # quat_mat = np.matrix(tr.quaternion_matrix(rot)) # dir = -(well_center - pos) # mag = np.linalg.norm(dir) # dir = dir / mag # mag = np.min(mag*10., 5) # print dir.T, mag # force_world = dir * mag # force_sensable = quat_mat[0:3, 0:3] * force_world # wr = Wrench() # wr.force.x = force_sensable[0] # wr.force.y = force_sensable[1] # wr.force.z = force_sensable[2] # wpub.publish(wr) #ps = PoseStamped() #ps.header.frame_id = 'world' #ps.header.stamp = rospy.get_rostime() #ps.pose = mat2pose(tip_world) #pub.publish(ps) #r.sleep() #Input force in link6's frame #Output force in omni base frame #wr = Wrench() #wr.force.x = np.random.rand()*2 - 1 #wr.force.y = np.random.rand()*2 - 1 #wr.force.z = np.random.rand()*2 - 1 #wpub.publish(wr)
[ [ 1, 0, 0.0071, 0.0071, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0142, 0.0071, 0, 0.66, 0.0476, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0213, 0.0071, 0, 0....
[ "import roslib", "roslib.load_manifest(\"phantom_omni\")", "import rospy", "import tf", "import tf.transformations as tr", "from geometry_msgs.msg import PoseStamped", "from geometry_msgs.msg import Pose", "from geometry_msgs.msg import Wrench", "import math", "import numpy as np", "def tf2mat(t...
import roslib roslib.load_manifest("phantom_omni") import rospy import tf import tf.transformations as tr from geometry_msgs.msg import PoseStamped from geometry_msgs.msg import Pose from geometry_msgs.msg import Wrench import math import numpy as np def tf2mat(tf_trans): (trans, rot) = tf_trans return np.matrix(tr.translation_matrix(trans)) * np.matrix(tr.quaternion_matrix(rot)) def mat2pose(m): trans = tr.translation_from_matrix(m) rot = tr.quaternion_from_matrix(m) p = Pose() p.position.x = trans[0] p.position.y = trans[1] p.position.z = trans[2] p.orientation.x = rot[0] p.orientation.y = rot[1] p.orientation.z = rot[2] p.orientation.w = rot[3] return p rospy.init_node("omni_potential_well") wpub = rospy.Publisher('force_feedback', Wrench) r = rospy.Rate(1000) #listener = tf.TransformListener() #listener.waitForTransform("/world", "/omni1_link6", rospy.Time(), rospy.Duration(4.0)) #listener.waitForTransform("/world", "/sensable", rospy.Time(), rospy.Duration(4.0)) print 'running.' well_center = None angle = 0.; while not rospy.is_shutdown(): #w_T_6 = tf2mat(listener.lookupTransform('/world', '/omni1_link6', rospy.Time(0))) #qm = np.matrix(tr.quaternion_matrix(tr.quaternion_from_euler(0, math.pi, 0))) #tm = np.matrix(tr.translation_matrix([-.134, 0, 0])) #tip_6 = tm * qm #tip_world = w_T_6 * tip_6 #pos = np.matrix(tr.translation_from_matrix(tip_world)).T #force_world = (w_T_6[0:3,0:3] * np.matrix([-2., 0, 0]).T) #trans, rot = listener.lookupTransform('/sensable', '/world', rospy.Time(0)) #quat_mat = np.matrix(tr.quaternion_matrix(rot)) #force_sensable = quat_mat[0:3, 0:3] * force_world wr = Wrench() angle = np.radians(.3) + angle wr.force.x = 0#force_sensable[0] wr.force.y = 0#force_sensable[1] wr.force.z = np.sin(angle) * 3#force_sensable[2] print wr.force.z wpub.publish(wr) r.sleep() #ps = PoseStamped() #ps.header.frame_id = 'sensable' #ps.header.stamp = rospy.get_rostime() #ps.pose = mat2pose(tip_world) #pub.publish(ps) #print force_sensable.T #if well_center == None: # well_center = pos #else: # trans, rot = listener.lookupTransform('/world', '/sensable', rospy.Time(0)) # quat_mat = np.matrix(tr.quaternion_matrix(rot)) # dir = -(well_center - pos) # mag = np.linalg.norm(dir) # dir = dir / mag # mag = np.min(mag*10., 5) # print dir.T, mag # force_world = dir * mag # force_sensable = quat_mat[0:3, 0:3] * force_world # wr = Wrench() # wr.force.x = force_sensable[0] # wr.force.y = force_sensable[1] # wr.force.z = force_sensable[2] # wpub.publish(wr) #ps = PoseStamped() #ps.header.frame_id = 'world' #ps.header.stamp = rospy.get_rostime() #ps.pose = mat2pose(tip_world) #pub.publish(ps) #r.sleep() #Input force in link6's frame #Output force in omni base frame #wr = Wrench() #wr.force.x = np.random.rand()*2 - 1 #wr.force.y = np.random.rand()*2 - 1 #wr.force.z = np.random.rand()*2 - 1 #wpub.publish(wr)
[ [ 1, 0, 0.0059, 0.0059, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0118, 0.0059, 0, 0.66, 0.0556, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0178, 0.0059, 0, 0....
[ "import roslib", "roslib.load_manifest(\"phantom_omni\")", "import rospy", "import tf", "import tf.transformations as tr", "from geometry_msgs.msg import PoseStamped", "from geometry_msgs.msg import Pose", "from geometry_msgs.msg import Wrench", "import math", "import numpy as np", "def tf2mat(t...
import roslib roslib.load_manifest("phantom_omni") import rospy import tf import tf.transformations as tr from geometry_msgs.msg import PoseStamped from geometry_msgs.msg import Pose from geometry_msgs.msg import Wrench import math import numpy as np def tf2mat(tf_trans): (trans, rot) = tf_trans return np.matrix(tr.translation_matrix(trans)) * np.matrix(tr.quaternion_matrix(rot)) def mat2pose(m): trans = tr.translation_from_matrix(m) rot = tr.quaternion_from_matrix(m) p = Pose() p.position.x = trans[0] p.position.y = trans[1] p.position.z = trans[2] p.orientation.x = rot[0] p.orientation.y = rot[1] p.orientation.z = rot[2] p.orientation.w = rot[3] return p rospy.init_node("pr2_force_feedback") pub = rospy.Publisher('pr2_master', PoseStamped) wpub = rospy.Publisher('force_feedback', Wrench) r = rospy.Rate(60) listener = tf.TransformListener() listener.waitForTransform("/world", "/omni1_link6", rospy.Time(), rospy.Duration(4.0)) print 'running.' while not rospy.is_shutdown(): w_T_6 = tf2mat(listener.lookupTransform('/world', '/omni1_link6', rospy.Time(0))) qm = np.matrix(tr.quaternion_matrix(tr.quaternion_from_euler(0, math.pi, 0))) tm = np.matrix(tr.translation_matrix([-.134, 0, 0])) tip_6 = tm * qm tip_world = w_T_6 * tip_6 ps = PoseStamped() ps.header.frame_id = 'world' ps.header.stamp = rospy.get_rostime() ps.pose = mat2pose(tip_world) pub.publish(ps) r.sleep() #Input force in link6's frame #Output force in omni base frame #wr = Wrench() #wr.force.x = np.random.rand()*2 - 1 #wr.force.y = np.random.rand()*2 - 1 #wr.force.z = np.random.rand()*2 - 1 #wpub.publish(wr) wr = Wrench() wr.force.x = 0 wr.force.y = 0 wr.force.z = 0 wpub.publish(wr)
[ [ 1, 0, 0.0147, 0.0147, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0294, 0.0147, 0, 0.66, 0.0526, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0441, 0.0147, 0, 0....
[ "import roslib", "roslib.load_manifest(\"phantom_omni\")", "import rospy", "import tf", "import tf.transformations as tr", "from geometry_msgs.msg import PoseStamped", "from geometry_msgs.msg import Pose", "from geometry_msgs.msg import Wrench", "import math", "import numpy as np", "def tf2mat(t...
import roslib roslib.load_manifest("phantom_omni") import rospy import tf import tf.transformations as tr import hrl_lib.tf_utils as tfu from geometry_msgs.msg import PoseStamped from geometry_msgs.msg import Pose from geometry_msgs.msg import Wrench import math import numpy as np rospy.init_node("omni_potential_well") wpub = rospy.Publisher('omni1_force_feedback', Wrench) def pose_cb(ps): m_f, frame = tfu.posestamped_as_matrix(ps) m_o1 = tfu.transform('/omni1', frame, listener) * m_f ee_point = np.matrix(tr.translation_from_matrix(m_o1)).T center = np.matrix([-.10, 0, .30]).T dif = 30*(center - ee_point) #force_dir = dif / np.linalg.norm(dif) force_o1 = dif #force_dir * np.sum(np.power(dif, 2)) force_s = tfu.transform('/omni1_sensable', '/omni1', listener) * np.row_stack((force_o1, np.matrix([1.]))) print np.linalg.norm(center - ee_point) wr = Wrench() wr.force.x = force_s[0] wr.force.y = force_s[1] wr.force.z = force_s[2] wpub.publish(wr) rospy.Subscriber('/omni1_pose', PoseStamped, pose_cb) r = rospy.Rate(1) listener = tf.TransformListener() listener.waitForTransform("/omni1", "/omni1_sensable", rospy.Time(), rospy.Duration(4.0)) listener.waitForTransform("/omni1", "/omni1_link6", rospy.Time(), rospy.Duration(4.0)) print 'running.' well_center = None while not rospy.is_shutdown(): r.sleep() #get current pose of the tip, subtract it from a center, rotate this into the sensable frame #tip_omni1 = w_T_6 * tip_6 #force_omni1 = (w_T_6[0:3,0:3] * np.matrix([-2., 0, 0]).T) #force_sensable = transform('/omni1_sensable', '/omni1', listener) * force_omni1 #wr = Wrench() #wr.force.x = force_sensable[0] #wr.force.y = force_sensable[1] #wr.force.z = force_sensable[2] #print wr.force.z #wpub.publish(wr) #qm = tfu.quaternion_from_matrix(tr.quaternion_from_euler(0, math.pi, 0)) #tm = tfu.translation_matrix([-.134, 0, 0]) #tip_6 = tm * qm #tip_omni1 = w_T_6 * tip_6 #trans, rot = listener.lookupTransform('/omni1_sensable', '/omni1', rospy.Time(0)) #quat_mat = np.matrix(tr.quaternion_matrix(rot)) #force_sensable = quat_mat[0:3, 0:3] * force_omni1 #def tf2mat(tf_trans): # (trans, rot) = tf_trans # return np.matrix(tr.translation_matrix(trans)) * np.matrix(tr.quaternion_matrix(rot)) #def mat2pose(m): # trans = tr.translation_from_matrix(m) # rot = tr.quaternion_from_matrix(m) # p = Pose() # p.position.x = trans[0] # p.position.y = trans[1] # p.position.z = trans[2] # p.orientation.x = rot[0] # p.orientation.y = rot[1] # p.orientation.z = rot[2] # p.orientation.w = rot[3] # return p
[ [ 1, 0, 0.009, 0.009, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.018, 0.009, 0, 0.66, 0.0476, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.027, 0.009, 0, 0.66, ...
[ "import roslib", "roslib.load_manifest(\"phantom_omni\")", "import rospy", "import tf", "import tf.transformations as tr", "import hrl_lib.tf_utils as tfu", "from geometry_msgs.msg import PoseStamped", "from geometry_msgs.msg import Pose", "from geometry_msgs.msg import Wrench", "import math", "...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('hrl_tilting_hokuyo') from enthought.mayavi import mlab import sys,os,time import optparse import hrl_lib.util as ut import numpy as np, math color_list = [(1.,1.,1.),(1.,0.,0.),(0.,1.,0.),(0.,0.,1.),(1.,1.,0.),(1.,0.,1.),\ (0.,1.,1.),(0.5,1.,0.5),(1.,0.5,0.5)] ## # make a figure with a white background. def white_bg(): mlab.figure(fgcolor = (0,0,0), bgcolor = (1,1,1)) ## # save plot as a png # @param name - file name # size - (r,c) e.g. (1024, 768) def savefig(name, size): mlab.savefig(name, size=size) ## plot 3D points connected to each other # # Check mlab.points3d documentation for details. # @param pts - 3xN numpy matrix of points. # @param color - 3 tuple of color. (float b/w 0 and 1) # @param mode - how to display the points ('point','sphere','cube' etc.) # @param scale_fator - controls size of the spheres. not sure what it means. def plot(pts,color=(1.,1.,1.), scalar_list=None): if scalar_list != None: mlab.plot3d(pts[0,:].A1,pts[1,:].A1,pts[2,:].A1,scalar_list, representation = 'wireframe', tube_radius = None) mlab.colorbar() else: mlab.plot3d(pts[0,:].A1,pts[1,:].A1,pts[2,:].A1,color=color, representation = 'wireframe', tube_radius = None) ## plot 3D points as a cloud. # # Check mlab.points3d documentation for details. # @param pts - 3xN numpy matrix of points. # @param color - 3 tuple of color. (float b/w 0 and 1) # @param mode - how to display the points ('point','sphere','cube' etc.) # @param scale_fator - controls size of the spheres. not sure what it means. def plot_points(pts,color=(1.,1.,1.),mode='point',scale_factor=0.01,scalar_list=None): if scalar_list != None: mlab.points3d(pts[0,:].A1,pts[1,:].A1,pts[2,:].A1,scalar_list,mode=mode,scale_factor=scale_factor) mlab.colorbar() else: mlab.points3d(pts[0,:].A1,pts[1,:].A1,pts[2,:].A1,mode=mode,color=color,scale_factor=scale_factor) ## Use mayavi2 to plot normals, and curvature of a point cloud. # @param pts - 3xN np matrix # @param normals - 3xN np matrix of surface normals at the points in pts. # @param curvature - list of curvatures. # @param mask_points - how many point to skip while drawint the normals # @param color - of the arrows # @param scale_factor - modulate size of arrows. # # Surface normals are plotted as arrows at the pts, curvature is colormapped and # shown as spheres. The radius of the sphere also indicates the magnitude # of the curvature. If curvature is None then it is not plotted. The pts # are then displayed as pixels. def plot_normals(pts, normals, curvature=None, mask_points=1, color=(0.,1.,0.), scale_factor = 0.1): x = pts[0,:].A1 y = pts[1,:].A1 z = pts[2,:].A1 u = normals[0,:].A1 v = normals[1,:].A1 w = normals[2,:].A1 if curvature != None: curvature = np.array(curvature) #idxs = np.where(curvature>0.03) #mlab.points3d(x[idxs],y[idxs],z[idxs],curvature[idxs],mode='sphere',scale_factor=0.1,mask_points=1) mlab.points3d(x,y,z,curvature,mode='sphere',scale_factor=0.1,mask_points=1, color=color) # mlab.points3d(x,y,z,mode='point') mlab.colorbar() else: mlab.points3d(x,y,z,mode='point') mlab.quiver3d(x, y, z, u, v, w, mask_points=mask_points, scale_factor=scale_factor, color=color) # mlab.axes() ## Plot a yellow cuboid. # cuboid is defined by 12 tuples of corners that define the 12 edges, # as returned by occupancy_grig.grid_lines() function. def plot_cuboid(corner_tups): for tup in corner_tups: p1 = tup[0] p2 = tup[1] mlab.plot3d([p1[0,0],p2[0,0]],[p1[1,0],p2[1,0]], [p1[2,0],p2[2,0]],color=(1.,1.,0.), representation='wireframe',tube_radius=None) ## show the plot. # call this function after plotting everything. def show(): mlab.show() if __name__ == '__main__': p = optparse.OptionParser() p.add_option('-c', action='store', type='string', dest='pts_pkl', help='pkl file with 3D points') p.add_option('-f', action='store', type='string', dest='dict_pkl', help='pkl file with 3D dict') p.add_option('--save_cloud', action='store_true', dest='save_cloud', help='pickle the point cloud (3xN matrix)') p.add_option('--pan_angle', action='store', type='float', dest='max_pan_angle', default=60.0, help='angle in DEGREES. points b/w (-pan_angle and +pan_angle) are displayed. [default=60.]') p.add_option('--max_dist', action='store', type='float', dest='max_dist', default=3.0, help='maximum distance (meters). Points further than this are discarded. [default=3.]') opt, args = p.parse_args() pts_pkl_fname = opt.pts_pkl dict_pkl_fname = opt.dict_pkl save_cloud_flag = opt.save_cloud max_pan_angle = opt.max_pan_angle max_dist = opt.max_dist if pts_pkl_fname != None: pts = ut.load_pickle(pts_pkl_fname) elif dict_pkl_fname != None: import tilting_hokuyo.processing_3d as p3d dict = ut.load_pickle(dict_pkl_fname) pts = p3d.generate_pointcloud(dict['pos_list'],dict['scan_list'], math.radians(-max_pan_angle), math.radians(max_pan_angle), dict['l1'],dict['l2'], min_tilt=math.radians(-90),max_tilt=math.radians(90)) else: print 'Specify either a pts pkl or a dict pkl (-c or -f)' print 'Exiting...' sys.exit() dist_mat = ut.norm(pts) idxs = np.where(dist_mat<max_dist)[1] print 'pts.shape', pts.shape pts = pts[:,idxs.A1] print 'pts.shape', pts.shape if save_cloud_flag: ut.save_pickle(pts,'numpy_pc_'+ut.formatted_time()+'.pkl') plot_points(pts) mlab.show()
[ [ 1, 0, 0.1562, 0.0052, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.1562, 0.0052, 0, 0.66, 0.0667, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.1667, 0.0052, 0, 0....
[ "import roslib; roslib.load_manifest('hrl_tilting_hokuyo')", "import roslib; roslib.load_manifest('hrl_tilting_hokuyo')", "from enthought.mayavi import mlab", "import sys,os,time", "import optparse", "import hrl_lib.util as ut", "import numpy as np, math", "color_list = [(1.,1.,1.),(1.,0.,0.),(0.,1.,0...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('hrl_tilting_hokuyo') import time import sys, optparse import numpy as np, math import hrl_lib.util as ut import robotis.robotis_servo as rs import hrl_hokuyo.hokuyo_scan as hs class tilt_hokuyo(): def __init__(self, dev_name, servo_id, hokuyo, baudrate=57600, l1=0.,l2=0.035, camera_name=None): ''' dev_name - name of serial device of the servo controller (e.g. '/dev/robot/servo0') servo_id - 2,3,4 ... (2 to 253) hokuyo - Hokuyo object. baudrate - for the servo controller. camera_name - name of the ''' self.servo = rs.robotis_servo(dev_name,servo_id,baudrate) self.hokuyo = hokuyo self.l1 = l1 self.l2 = l2 def scan(self, range, speed, save_scan=False,avg=1, _max_retries=0): ''' range - (start,end) in radians speed - scan speed in radians/sec save_scan - save a dict of pos_list,scan_list,l1,l2 avg - average scans from the hokuyo. returns pos_list,scan_list. list of angles and HokuyoScans ''' ramp_up_angle = math.radians(5) if abs(range[0])+ramp_up_angle > math.radians(95) or \ abs(range[1])+ramp_up_angle > math.radians(95): print 'tilt_hokuyo_servo.scan:bad angles- ',math.degrees(range[0]),math.degrees(range[1]) min_angle = min(range[0],range[1]) max_angle = max(range[0],range[1]) # if max_angle>math.radians(60.5): # print 'tilt_hokuyo_servo.scan: maximum angle is too high, will graze bottom plate of mount. angle:', math.degrees(max_angle) # sys.exit() self.servo.move_angle(range[0]+np.sign(range[0])*ramp_up_angle) # time.sleep(0.05) # while(self.servo.is_moving()): # continue self.servo.move_angle(range[1]+np.sign(range[1])*ramp_up_angle,speed,blocking=False) #self.servo.move_angle(range[1], speed) time.sleep(0.05) t1 = time.time() pos_list = [] scan_list = [] while self.servo.is_moving(): pos = self.servo.read_angle() #print 'h6', pos if pos < min_angle or pos > max_angle: continue pos_list.append(pos) plane_scan = self.hokuyo.get_scan(avoid_duplicate=True,remove_graze=True,avg=avg) scan_list.append(plane_scan) t2 = time.time() self.servo.move_angle(0) if save_scan: date_name = ut.formatted_time() dict = {'pos_list': pos_list,'scan_list': scan_list, 'l1': self.l1, 'l2': self.l2} ut.save_pickle(dict,date_name+'_dict.pkl') runtime = t2 - t1 expected_number_scans = 19.0 * runtime * (1.0/avg) scan_threshold = expected_number_scans - expected_number_scans*.2 if len(scan_list) < scan_threshold: print 'tilt_hokuyo_servo.scan: WARNING! Expected at least %d scans but got only %d scans.' % (expected_number_scans, len(scan_list)) print 'tilt_hokuyo_servo.scan: trying again.. retries:', _max_retries if _max_retries > 0: return self.scan(range, speed, save_scan, avg, _max_retries = _max_retries-1) else: print 'tilt_hokuyo_servo.scan: returning anyway' print 'tilt_hokuyo_servo.scan: got %d scans over range %f with speed %f.' % (len(scan_list), (max_angle - min_angle), speed) return pos_list,scan_list def scan_around_pt(self,pt,speed=math.radians(5)): ''' pt - in thok coord frame. this function scans in a fixed range. returns pos_lit,scan_list ''' ang1 = math.radians(40) ang2 = math.radians(0) tilt_angles = (ang1,ang2) pos_list,scan_list = self.scan(tilt_angles,speed=speed) return pos_list,scan_list if __name__ == '__main__': # urg mount - l1=0.06, l2=0.05 # utm - l1 = 0.0, l2 = 0.035 p = optparse.OptionParser() p.add_option('-d', action='store', type='string', dest='servo_dev_name', default='/dev/robot/servo0', help='servo device string. [default= /dev/robot/servo0]') p.add_option('-t', action='store', type='string', dest='hokuyo_type',default='utm', help='hokuyo_type. urg or utm [default=utm]') p.add_option('-n', action='store', type='int', dest='hokuyo_number', default=0, help='hokuyo number. 0,1,2 ... [default=0]') p.add_option('--save_scan',action='store_true',dest='save_scan', help='save the scan [dict and cloud]') p.add_option('--speed', action='store', type='float', dest='scan_speed', help='scan speed in deg/s.[default=5]',default=5.) p.add_option('--ang0', action='store', type='float', dest='ang0', help='starting tilt angle for scan (degrees). [default=20]', default=20.0) p.add_option('--ang1', action='store', type='float', dest='ang1', help='ending tilt angle for scan (degrees). default=[-20]', default=-20.0) p.add_option('--id', action='store', type='int', dest='servo_id', default=2, help='servo id 1,2 ... [default=2]') p.add_option('--l2', action='store', type='float', dest='l2', help='l2 (in meters) [0.035 for ElE, -0.055 for mekabot]') p.add_option('--flip', action='store_true', dest='flip',help='flip the hokuyo scan') opt, args = p.parse_args() hokuyo_type = opt.hokuyo_type hokuyo_number = opt.hokuyo_number servo_dev_name = opt.servo_dev_name save_scan = opt.save_scan scan_speed = math.radians(opt.scan_speed) ang0 = opt.ang0 ang1 = opt.ang1 servo_id = opt.servo_id l2 = opt.l2 if l2==None: print 'please specify l2. do -h for details.' print 'Exiting...' sys.exit() flip = opt.flip if hokuyo_type == 'utm': h = hs.Hokuyo('utm',hokuyo_number,flip=flip) elif hokuyo_type == 'urg': h = hs.Hokuyo('urg',hokuyo_number,flip=flip) else: print 'unknown hokuyo type: ', hokuyo_type thok = tilt_hokuyo(servo_dev_name,servo_id,h,l1=0.,l2=l2) tilt_angles = (math.radians(ang0),math.radians(ang1)) pos_list,scan_list = thok.scan(tilt_angles,speed=scan_speed,save_scan=save_scan) sys.exit() # to kill the hokuyo thread.
[ [ 1, 0, 0.1571, 0.0052, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.1571, 0.0052, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.1675, 0.0052, 0, 0....
[ "import roslib; roslib.load_manifest('hrl_tilting_hokuyo')", "import roslib; roslib.load_manifest('hrl_tilting_hokuyo')", "import time", "import sys, optparse", "import numpy as np, math", "import hrl_lib.util as ut", "import robotis.robotis_servo as rs", "import hrl_hokuyo.hokuyo_scan as hs", "clas...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('hrl_tilting_hokuyo') import hrl_hokuyo.hokuyo_processing as hp import sys, optparse, os import hrl_lib.util as ut import hrl_lib.transforms as tr import numpy as np,math import time import display_3d_mayavi as d3m import occupancy_grid_3d as og3d import scipy.ndimage as ni import copy import pylab as pl color_list = [(1.,1.,0),(1.,0,0),(0,1.,1.),(0,1.,0),(0,0,1.),(0,0.4,0.4),(0.4,0.4,0), (0.4,0,0.4),(0.4,0.8,0.4),(0.8,0.4,0.4),(0.4,0.4,0.8),(0.4,0,0.8),(0,0.8,0.4), (0,0.4,0.8),(0.8,0,0.4),(0.4,0,0.4),(1.,0.6,0.02) ] #--------------------- utility functions ------------- ##returns points that are within the cuboid formed by tlb and brf. # @param pts - 3xN numpy matrix # @param tlb, brf - 3x1 np matrix. bottom-right-front and top-left-back # @return 3xM numpy matrix def points_within_cuboid(pts,brf,tlb): idx = np.where(np.min(np.multiply(pts>brf,pts<tlb),0))[1] if idx.shape[1] == 0: within_pts = np.empty((3,0)) else: within_pts = pts[:,idx.A1.tolist()] return within_pts def generate_pointcloud(pos_list, scan_list, min_angle, max_angle, l1, l2,save_scan=False, max_dist=np.Inf, min_dist=-np.Inf,min_tilt=-np.Inf,max_tilt=np.Inf, get_intensities=False, reject_zero_ten=True): ''' pos_list - list of motor positions (in steps) scan_list - list of UrgScan objects at the corres positions. l1,l2 - parameterizing the transformation (doc/ folder) save_scan - pickle 3xN numpy matrix of points. max_dist - ignore points at distance > max_dist min_dist - ignore points at distance < max_dist min_tilt, max_tilt - min and max tilt angles (radians) +ve tilts the hokuyo down. get_intensites - additionally return intensity values returns 3xN numpy matrix of 3d coords of the points, returns (3xN, 1xN) including the intensity values if get_intensites=True ''' t0 = time.time() allpts = [] allintensities = [] pos_arr = np.array(pos_list) scan_arr = np.array(scan_list) idxs = np.where(np.multiply(pos_arr<=max_tilt,pos_arr>=min_tilt)) pos_list = pos_arr[idxs].tolist() scan_list = scan_arr[idxs].tolist() n_scans = len(scan_list) if n_scans>1: scan_list = copy.deepcopy(scan_list) # remove graze in across scans. ranges_mat = [] for s in scan_list: ranges_mat.append(s.ranges.T) ranges_mat = np.column_stack(ranges_mat) start_index = hp.angle_to_index(scan_list[0],min_angle) end_index = hp.angle_to_index(scan_list[0],max_angle) for r in ranges_mat[start_index:end_index+1]: hp.remove_graze_effect(r,np.matrix(pos_list),skip=1,graze_angle_threshold=math.radians(169.)) for i,s in enumerate(scan_list): s.ranges = ranges_mat[:,i].T for p,s in zip(pos_list,scan_list): mapxydata = hp.get_xy_map(s,min_angle,max_angle,max_dist=max_dist,min_dist=min_dist,reject_zero_ten=reject_zero_ten,sigmoid_hack=True,get_intensities=get_intensities) # pts is 2xN if get_intensities == True: pts, intensities = mapxydata allintensities.append(intensities) else: pts = mapxydata pts = np.row_stack((pts,np.zeros(pts.shape[1]))) # pts is now 3xN pts = tr.Ry(-p)*(pts+np.matrix((l1,0,-l2)).T) allpts.append(pts) allpts = np.column_stack(allpts) if save_scan: date_name = ut.formatted_time() ut.save_pickle(allpts,date_name+'_cloud.pkl') t1 = time.time() # print 'Time to generate point cloud:', t1-t0 # allpts = tr.rotZ(math.radians(5))*allpts if get_intensities == True: allintensities = np.column_stack(allintensities) return allpts, allintensities else: return allpts #----------------------- navigation functions ----------------- def max_fwd_without_collision(all_pts,z_height,max_dist,display_list=None): ''' find the max distance that it is possible to move fwd by without collision. all_pts - 3xN matrix of 3D points in thok frame. z - height of zenither while taking the scan. max_dist - how far do I want to check for a possible collision. returns max_dist that the thok can be moved fwd without collision. ''' brf = np.matrix([0.2,-0.4,-z_height-0.1]).T tlb = np.matrix([max_dist, 0.4,-z_height+1.8]).T resolution = np.matrix([0.05,0.05,0.02]).T gr = og3d.occupancy_grid_3d(brf,tlb,resolution) gr.fill_grid(all_pts) gr.to_binary(4) ground_z = tr.thok0Tglobal(np.matrix([0,0,-z_height]).T)[2,0] # gr.remove_horizontal_plane(hmax=ground_z+0.1) gr.remove_horizontal_plane(extra_layers=2) collide_pts = np.row_stack(np.where(gr.grid!=0)) x_coords = collide_pts[0] # print x_coords if x_coords.shape[0] == 0: max_x = max_dist # height_mat = np.arrary([np.Inf]) else: max_x_gr = np.min(x_coords) # height_mat = collide_pts[1,np.where(x_coords==max_x_gr)[0]] # height_mat = height_mat*gr.resolution[2,0]+gr.brf[2,0] max_x = max_x_gr*gr.resolution[0,0]+gr.brf[0,0] if display_list != None: collide_grid = gr.grid_to_points() display_list.append(pu.PointCloud(all_pts,(200,200,200))) display_list.append(pu.CubeCloud(collide_grid,(200,200,200),resolution)) display_list.append(pu.CubeCloud(np.matrix((max_x,0.,0.)).T,(0,0,200),size=np.matrix([0.05,.05,0.05]).T)) # return max_x,np.matrix(height_mat) return max_x def find_goto_point_surface_1(grid,pt,display_list=None): ''' returns p_erratic,p_edge,surface_height p_erratic - point where the erratic's origin should go to. (in thok0 frame) p_edge - point on the edge closest to pt. p_erratic,p_edge are 3x1 matrices. surface_height - in thok0 coord frame. ''' pt_thok = pt # everything is happening in the thok coord frame. close_pt,approach_vector = find_approach_direction(grid,pt_thok,display_list) if close_pt == None: return None,None,None # move perpendicular to approach direction. # lam = -(close_pt[0:2,0].T*approach_vector)[0,0] # lam = min(lam,-0.3) # atleast 0.3m from the edge. lam = -0.4 goto_pt = close_pt[0:2,0] + lam*approach_vector # this is where I want the utm # to be err_to_thok = tr.erraticTglobal(tr.globalTthok0(np.matrix([0,0,0]).T)) goto_pt_erratic = -err_to_thok[0,0]*approach_vector + goto_pt # this is NOT # general. It uses info about the two frames. If frames move, bad # things can happen. if display_list != None: display_list.append(pu.CubeCloud(np.row_stack((goto_pt,close_pt[2,0])),color=(0,250,250), size=(0.012,0.012,0.012))) display_list.append(pu.Line(np.row_stack((goto_pt,close_pt[2,0])).A1,close_pt.A1,color=(255,20,0))) p_erratic = np.row_stack((goto_pt_erratic,np.matrix((close_pt[2,0])))) print 'p_erratic in thok0:', p_erratic.T return p_erratic,close_pt,close_pt[2,0] #------------------ surface orientation ---------------- def find_closest_pt_index(pts2d,pt): ''' returns index (of pts2d) of closest point to pt. pts2d - 2xN matrix, pt - 2x1 matrix ''' pt_to_edge_dists = ut.norm(pts2d-pt) closest_pt_index = np.argmin(pt_to_edge_dists) return closest_pt_index def find_closest_pt(pts2d,pt,pt_closer=False): ''' returns closest point to edge (2x1 matrix) can return None also ''' dist_pt = np.linalg.norm(pt[0:2,0]) pts2d_r = ut.norm(pts2d) pts2d_a = np.arctan2(pts2d[1,:],pts2d[0,:]) if pt_closer == False: k_idxs = np.where(pts2d_r<=dist_pt) else: k_idxs = np.where(pts2d_r>dist_pt) pts2d_r = pts2d_r[k_idxs] pts2d_a = pts2d_a[k_idxs] pts2d = ut.cart_of_pol(np.matrix(np.row_stack((pts2d_r,pts2d_a)))) if pt_closer == False: edge_to_pt = pt[0:2,0]-pts2d else: edge_to_pt = pts2d-pt[0:2,0] edge_to_pt_a = np.arctan2(edge_to_pt[1,:],edge_to_pt[0,:]) keep_idxs = np.where(np.abs(edge_to_pt_a)<math.radians(70))[1].A1 if keep_idxs.shape[0] == 0: return None pts2d = pts2d[:,keep_idxs] # pt_to_edge_dists = ut.norm(pts2d-pt[0:2,0]) # closest_pt_index = np.argmin(pt_to_edge_dists) closest_pt_index = find_closest_pt_index(pts2d,pt[0:2,0]) closest_pt = pts2d[:,closest_pt_index] return closest_pt def pushback_edge(pts2d,pt): ''' push pt away from the edge defined by pts2d. pt - 2x1, pts2d - 2xN returns the pushed point. ''' closest_idx = find_closest_pt_index(pts2d,pt) n_push_points = min(min(5,pts2d.shape[1]-closest_idx-1),closest_idx) if closest_idx<n_push_points or (pts2d.shape[1]-closest_idx-1)<n_push_points: print 'processing_3d.pushback_edge: pt is too close to the ends of the pts2d array.' return None edge_to_pt = pt-pts2d[:,closest_idx-n_push_points:closest_idx+n_push_points] edge_to_pt_r = ut.norm(edge_to_pt) edge_to_pt_a = np.arctan2(edge_to_pt[1,:],edge_to_pt[0,:]) non_zero_idxs = np.where(edge_to_pt_r>0.005) edge_to_pt_r = edge_to_pt_r[non_zero_idxs] edge_to_pt_r[0,:] = 1 edge_to_pt_a = edge_to_pt_a[non_zero_idxs] edge_to_pt_unit = ut.cart_of_pol(np.row_stack((edge_to_pt_r,edge_to_pt_a))) push_vector = edge_to_pt_unit.mean(1) push_vector = push_vector/np.linalg.norm(push_vector) print 'push_vector:', push_vector.T pt_pushed = pt + push_vector*0.05 return pt_pushed ## figure out a good direction to approach the surface. # @param grid - occupancy grid (binary) around the point of interest. # assumes that it has a surface. # @param pt - 3d point which has to be approached. # @param display_list - if display_list are lists then point clouds etc. are added # for visualisation. # # @return - closest_pt,approach_vector. def find_approach_direction(grid,pt,display_list=None): z_plane,max_count = grid.argmax_z(search_up=True) z_plane_meters = z_plane*grid.resolution[2,0]+grid.brf[2,0] l = grid.find_plane_indices(assume_plane=True) print '------------ min(l)',min(l) z_plane_argmax,max_count = grid.argmax_z(search_up=False) z_plane_below = max(0,z_plane_argmax-5) print 'z_plane_argmax',z_plane_argmax print 'z_plane_below',z_plane_below print 'l:',l # l = range(z_plane_below,z_plane)+l copy_grid = copy.deepcopy(grid) plane_slices = grid.grid[:,:,l] copy_grid.grid[:,:,:] = 0 copy_grid.grid[:,:,l] = copy.copy(plane_slices) #display_list.append(pu.PointCloud(copy_grid.grid_to_points(),color=(0,0,255))) #plane_pts = copy_grid.grid_to_points() grid_2d = np.max(grid.grid[:,:,l],2) grid_2d = ni.binary_dilation(grid_2d,iterations=4) # I want 4-connectivity while filling holes. grid_2d = ni.binary_fill_holes(grid_2d) # I want 4-connectivity while filling holes. labeled_arr,n_labels = ni.label(grid_2d) labels_list = range(1,n_labels+1) count_objects = ni.sum(grid_2d,labeled_arr,labels_list) if n_labels == 1: count_objects = [count_objects] max_label = np.argmax(np.matrix(count_objects)) grid_2d[np.where(labeled_arr!=max_label+1)] = 0 # connect_structure = np.empty((3,3),dtype='int') # connect_structure[:,:] = 1 # eroded_2d = ni.binary_erosion(grid_2d,connect_structure,iterations=4) # eroded_2d = ni.binary_erosion(grid_2d) # grid_2d = grid_2d-eroded_2d labeled_arr_3d = copy_grid.grid.swapaxes(2,0) labeled_arr_3d = labeled_arr_3d.swapaxes(1,2) labeled_arr_3d = labeled_arr_3d*grid_2d labeled_arr_3d = labeled_arr_3d.swapaxes(2,0) labeled_arr_3d = labeled_arr_3d.swapaxes(1,0) copy_grid.grid = labeled_arr_3d pts3d = copy_grid.grid_to_points() pts2d = pts3d[0:2,:] dist_pt = np.linalg.norm(pt[0:2,0]) pts2d_r = ut.norm(pts2d) pts2d_a = np.arctan2(pts2d[1,:],pts2d[0,:]) max_angle = np.max(pts2d_a) min_angle = np.min(pts2d_a) max_angle = min(max_angle,math.radians(50)) min_angle = max(min_angle,math.radians(-50)) ang_res = math.radians(1.) n_bins = int((max_angle-min_angle)/ang_res) print 'n_bins:', n_bins n_bins = min(50,n_bins) # n_bins=50 ang_res = (max_angle-min_angle)/n_bins print 'n_bins:', n_bins angle_list = [] range_list = [] for i in range(n_bins): angle = min_angle+ang_res*i idxs = np.where(np.multiply(pts2d_a<(angle+ang_res/2.),pts2d_a>(angle-ang_res/2.))) r_mat = pts2d_r[idxs] a_mat = pts2d_a[idxs] if r_mat.shape[1] == 0: continue min_idx = np.argmin(r_mat.A1) range_list.append(r_mat[0,min_idx]) angle_list.append(a_mat[0,min_idx]) if range_list == []: print 'processing_3d.find_approach_direction: No edge points found' return None,None pts2d = ut.cart_of_pol(np.matrix(np.row_stack((range_list,angle_list)))) closest_pt_1 = find_closest_pt(pts2d,pt,pt_closer=False) if closest_pt_1 == None: dist1 = np.Inf else: approach_vector_1 = pt[0:2,0] - closest_pt_1 dist1 = np.linalg.norm(approach_vector_1) approach_vector_1 = approach_vector_1/dist1 closest_pt_2 = find_closest_pt(pts2d,pt,pt_closer=True) if closest_pt_2 == None: dist2 = np.Inf else: approach_vector_2 = closest_pt_2 - pt[0:2,0] dist2 = np.linalg.norm(approach_vector_2) approach_vector_2 = approach_vector_2/dist2 if dist1 == np.Inf and dist2 == np.Inf: approach_vector_1 = np.matrix([1.,0.,0.]).T approach_vector_2 = np.matrix([1.,0.,0.]).T print 'VERY STRANGE: processing_3d.find_approach_direction: both distances are Inf' if dist1<0.05 or dist2<0.05: print 'dist1,dist2:',dist1,dist2 t_pt = copy.copy(pt) if dist1<dist2 and dist1<0.02: t_pt[0,0] += 0.05 elif dist2<0.02: t_pt[0,0] -= 0.05 #pt_new = pushback_edge(pts2d,pt[0:2,0]) pt_new = pushback_edge(pts2d,t_pt[0:2,0]) if display_list != None: pt_new_3d = np.row_stack((pt_new,np.matrix([z_plane_meters]))) display_list.append(pu.CubeCloud(pt_new_3d,color=(200,000,0),size=(0.009,0.009,0.009))) closest_pt_1 = find_closest_pt(pts2d,pt_new,pt_closer=False) if closest_pt_1 == None: dist1 = np.Inf else: approach_vector_1 = pt_new - closest_pt_1 dist1 = np.linalg.norm(approach_vector_1) approach_vector_1 = approach_vector_1/dist1 closest_pt_2 = find_closest_pt(pts2d,pt_new,pt_closer=True) if closest_pt_2 == None: dist2 = np.Inf else: approach_vector_2 = closest_pt_2 - pt_new dist2 = np.linalg.norm(approach_vector_2) approach_vector_2 = approach_vector_2/dist2 print '----------- dist1,dist2:',dist1,dist2 if dist2<dist1: closest_pt = closest_pt_2 approach_vector = approach_vector_2 else: closest_pt = closest_pt_1 approach_vector = approach_vector_1 print '----------- approach_vector:',approach_vector.T closest_pt = np.row_stack((closest_pt,np.matrix([z_plane_meters]))) if display_list != None: z = np.matrix(np.empty((1,pts2d.shape[1]))) z[:,:] = z_plane_meters pts3d_front = np.row_stack((pts2d,z)) display_list.append(pu.CubeCloud(closest_pt,color=(255,255,0),size=(0.020,0.020,0.020))) display_list.append(pu.CubeCloud(pts3d_front,color=(255,0,255),size=grid.resolution)) #display_list.append(pu.CubeCloud(pts3d,color=(0,255,0))) return closest_pt,approach_vector #------------------- for doors --------------------- def vertical_plane_points(grid): ''' changes grid ''' plane_indices,ver_plane_slice = grid.remove_vertical_plane() grid.grid[:,:,:] = 0 grid.grid[plane_indices,:,:] = ver_plane_slice ## returns door handle points in the thok coord frame. def find_door_handle(grid,pt,list = None,rotation_angle=math.radians(0.), occupancy_threshold=None,resolution=None): grid.remove_vertical_plane() pts = grid.grid_to_points() rot_mat = tr.Rz(rotation_angle) t_pt = rot_mat*pt brf = t_pt+np.matrix([-0.2,-0.3,-0.2]).T tlb = t_pt+np.matrix([0.2, 0.3,0.2]).T #resolution = np.matrix([0.02,0.0025,0.02]).T grid = og3d.occupancy_grid_3d(brf,tlb,resolution,rotation_z=rotation_angle) if pts.shape[1] == 0: return None grid.fill_grid(tr.Rz(rotation_angle)*pts) grid.to_binary(occupancy_threshold) labeled_arr,n_labels = grid.find_objects() if list == None: object_points_list = [] else: object_points_list = list for l in range(n_labels): pts = grid.labeled_array_to_points(labeled_arr,l+1) obj_height = np.max(pts[2,:])-np.min(pts[2,:]) print 'object_height:', obj_height if obj_height > 0.1: #remove the big objects grid.grid[np.where(labeled_arr==l+1)] = 0 connect_structure = np.empty((3,3,3),dtype='int') connect_structure[:,:,:] = 0 connect_structure[1,:,1] = 1 # dilate away - actual width of the door handle is not affected # because that I will get from the actual point cloud! grid.grid = ni.binary_dilation(grid.grid,connect_structure,iterations=7) labeled_arr,n_labels = grid.find_objects() for l in range(n_labels): pts = grid.labeled_array_to_points(labeled_arr,l+1) pts2d = pts[1:3,:] # only the y-z coordinates. obj_width = (pts2d.max(1)-pts2d.min(1))[0,0] print 'processing_3d.find_door_handle: object width = ', obj_width if obj_width < 0.05: continue pts2d_zeromean = pts2d-pts2d.mean(1) e_vals,e_vecs = np.linalg.eig(pts2d_zeromean*pts2d_zeromean.T) max_index = np.argmax(e_vals) max_evec = e_vecs[:,max_index] ang = math.atan2(max_evec[1,0],max_evec[0,0]) print 'processing_3d.find_door_handle: ang = ', math.degrees(ang) if (ang>math.radians(45) and ang<math.radians(135)) or \ (ang>math.radians(-135) and ang<math.radians(-45)): # assumption is that door handles are horizontal. continue object_points_list.append(pts) print 'processing_3d.find_door_handle: found %d objects'%(len(object_points_list)) closest_obj = find_closest_object(object_points_list,pt) return closest_obj #--------------------- segmentation --------------------- def find_closest_object(obj_pts_list,pt,return_idx=False): ''' obj_pts_list - list of 3xNi matrices of points. pt - point of interest. (3x1) matrix. return_idx - whether to return the index (in obj_pts_list) of the closest object. returns 3xNj matrix of points which is the closest object to pt. None if obj_pts_list is empty. ''' min_dist_list = [] for obj_pts in obj_pts_list: min_dist_list.append(np.min(ut.norm(obj_pts-pt))) if obj_pts_list == []: return None min_idx = np.argmin(np.matrix(min_dist_list)) cl_obj = obj_pts_list[min_idx] print 'processing_3d.find_closest_object: closest_object\'s centroid',cl_obj.mean(1).T if return_idx: return cl_obj,min_idx return cl_obj def segment_objects_points(grid,return_labels_list=False, twod=False): ''' grid - binary occupancy grid. returns list of 3xNi numpy matrices where Ni is the number of points in the ith object. Point refers to center of the cell of occupancy grid. return_labels_list - return a list of labels of the objects in the grid. returns None if there is no horizontal surface ''' labeled_arr,n_labels = grid.segment_objects(twod=twod) if n_labels == None: # there is no surface, so segmentation does not make sense. return None object_points_list = [] labels_list = [] for l in range(n_labels): pts = grid.labeled_array_to_points(labeled_arr,l+1) pts_zeromean = pts-pts.mean(1) e_vals,e_vecs = np.linalg.eig(pts_zeromean*pts_zeromean.T) max_index = np.argmax(e_vals) max_evec = e_vecs[:,max_index] print 'max eigen vector:', max_evec.T pts_1d = max_evec.T * pts size = pts_1d.max() - pts_1d.min() print 'size:', size print 'n_points:', pts.shape[1] ppsoe = pts.shape[1]/(e_vals[0]+e_vals[1]+e_vals[2]) print 'points per sum of eigenvalues:',ppsoe # if ppsoe<5000: # continue if size<0.05 or size>0.5: #TODO - figure out a good threshold. continue object_points_list.append(pts) labels_list.append(l+1) if return_labels_list: return object_points_list, labels_list return object_points_list def create_grid(brf,tlb,resolution,pos_list,scan_list,l1,l2, display_flag=False,show_pts=True,rotation_angle=0., occupancy_threshold=1): ''' rotation angle - about the Z axis. ''' max_dist = np.linalg.norm(tlb) + 0.2 min_dist = brf[0,0] min_angle,max_angle=math.radians(-60),math.radians(60) all_pts = generate_pointcloud(pos_list, scan_list, min_angle, max_angle, l1, l2, max_dist=max_dist,min_dist=min_dist) rot_mat = tr.Rz(rotation_angle) all_pts_rot = rot_mat*all_pts gr = og3d.occupancy_grid_3d(brf,tlb,resolution,rotation_z=rotation_angle) gr.fill_grid(all_pts_rot) gr.to_binary(occupancy_threshold) if display_flag == True: if show_pts: d3m.plot_points(all_pts,color=(0.,0.,0.)) cube_tups = gr.grid_lines(rotation_angle=rotation_angle) d3m.plot_cuboid(cube_tups) return gr def create_vertical_plane_grid(pt,pos_list,scan_list,l1,l2,rotation_angle,display_list=None): rot_mat = tr.Rz(rotation_angle) t_pt = rot_mat*pt brf = t_pt+np.matrix([-0.2,-0.3,-0.2]).T tlb = t_pt+np.matrix([0.2, 0.3,0.2]).T resolution = np.matrix([0.005,0.02,0.02]).T return create_grid(brf,tlb,resolution,pos_list,scan_list,l1,l2,display_list,rotation_angle=rotation_angle,occupancy_threshold=1) def create_scooping_grid(pt,pos_list,scan_list,l1,l2,display_flag=False): brf = pt+np.matrix([-0.15,-0.4,-0.2]).T brf[0,0] = max(0.07,brf[0,0]) tlb = pt+np.matrix([0.25, 0.4,0.2]).T resolution = np.matrix([0.01,0.01,0.0025]).T return create_grid(brf,tlb,resolution,pos_list,scan_list,l1,l2,display_flag) def create_segmentation_grid(pt,pos_list,scan_list,l1,l2,display_flag=False): brf = pt+np.matrix([-0.15,-0.2,-0.2]).T brf[0,0] = max(0.07,brf[0,0]) tlb = pt+np.matrix([0.25, 0.2,0.2]).T # brf = np.matrix([0.05,-0.3,-0.2]).T # tlb = np.matrix([0.5, 0.3,0.1]).T # resolution = np.matrix([0.005,0.005,0.005]).T # resolution = np.matrix([0.005,0.005,0.0025]).T resolution = np.matrix([0.01,0.01,0.0025]).T # resolution = np.matrix([0.01,0.01,0.001]).T return create_grid(brf,tlb,resolution,pos_list,scan_list,l1,l2,display_flag) def create_approach_grid(pt,pos_list,scan_list,l1,l2,display_list=None,show_pts=True): brf = pt + np.matrix([-0.5,-0.2,-0.3]).T brf[0,0] = max(0.10,brf[0,0]) tlb = pt + np.matrix([0.3, 0.2,0.2]).T # resolution = np.matrix([0.005,0.005,0.005]).T resolution = np.matrix([0.01,0.01,0.0025]).T # resolution = np.matrix([0.01,0.01,0.001]).T return create_grid(brf,tlb,resolution,pos_list,scan_list,l1,l2,display_list,show_pts) def create_overhead_grasp_choice_grid(pt,pos_list,scan_list,l1,l2,far_dist,display_list=None): # y_pos = max(pt[1,0]+0.1,0.1) # y_neg = min(pt[1,0]-0.1,-0.1) # # brf = np.matrix([0.2,y_neg,0.0]).T # tlb = np.matrix([pt[0,0]+far_dist,y_pos,pt[2,0]+0.75]).T # # resolution = np.matrix([0.02,0.02,0.02]).T y_pos = 0.1 y_neg = -0.1 r = np.linalg.norm(pt[0:2,0]) brf = np.matrix([0.25,y_neg,0.0]).T tlb = np.matrix([r+far_dist,y_pos,pt[2,0]+0.75]).T resolution = np.matrix([0.02,0.02,0.02]).T rotation_angle = math.atan2(pt[1,0],pt[0,0]) print 'rotation_angle:', math.degrees(rotation_angle) gr = create_grid(brf,tlb,resolution,pos_list,scan_list,l1,l2,display_list,rotation_angle=rotation_angle) if display_list != None: collide_pts = gr.grid_to_points() if collide_pts.shape[1] > 0: collide_pts = tr.Rz(rotation_angle).T*gr.grid_to_points() display_list.insert(0,pu.PointCloud(collide_pts,color=(0,0,0))) return gr def overhead_grasp_collision(pt,grid): print 'collision points:', grid.grid.sum() if grid.grid.sum()>15: return True else: return False def grasp_location_on_object(obj,display_list=None): ''' obj - 3xN numpy matrix of points of the object. ''' pts_2d = obj[0:2,:] centroid_2d = pts_2d.mean(1) pts_2d_zeromean = pts_2d-centroid_2d e_vals,e_vecs = np.linalg.eig(pts_2d_zeromean*pts_2d_zeromean.T) # get the min size min_index = np.argmin(e_vals) min_evec = e_vecs[:,min_index] print 'min eigenvector:', min_evec.T pts_1d = min_evec.T * pts_2d min_size = pts_1d.max() - pts_1d.min() print 'spread along min eigenvector:', min_size max_height = obj[2,:].max() tlb = obj.max(1) brf = obj.min(1) print 'tlb:', tlb.T print 'brf:', brf.T resolution = np.matrix([0.005,0.005,0.005]).T gr = og3d.occupancy_grid_3d(brf,tlb,resolution) gr.fill_grid(obj) gr.to_binary(1) obj = gr.grid_to_points() grid_2d = gr.grid.max(2) grid_2d_filled = ni.binary_fill_holes(grid_2d) gr.grid[:,:,0] = gr.grid[:,:,0]+grid_2d_filled-grid_2d p = np.matrix(np.row_stack(np.where(grid_2d_filled==1))).astype('float') p[0,:] = p[0,:]*gr.resolution[0,0] p[1,:] = p[1,:]*gr.resolution[1,0] p += gr.brf[0:2,0] print 'new mean:', p.mean(1).T print 'centroid_2d:', centroid_2d.T centroid_2d = p.mean(1) if min_size<0.12: # grasp at centroid. grasp_point = np.row_stack((centroid_2d,np.matrix([max_height+gr.resolution[2,0]*2]))) # grasp_point[2,0] = max_height gripper_angle = -math.atan2(-min_evec[0,0],min_evec[1,0]) grasp_vec = min_evec if display_list != None: max_index = np.argmax(e_vals) max_evec = e_vecs[:,max_index] pts_1d = max_evec.T * pts_2d max_size = pts_1d.max() - pts_1d.min() v = np.row_stack((max_evec,np.matrix([0.]))) max_end_pt1 = grasp_point + v*max_size/2. max_end_pt2 = grasp_point - v*max_size/2. display_list.append(pu.Line(max_end_pt1,max_end_pt2,color=(0,0,0))) else: #----- more complicated grasping location finder. for i in range(gr.grid_shape[2,0]): gr.grid[:,:,i] = gr.grid[:,:,i]*(i+1) height_map = gr.grid.max(2) * gr.resolution[2,0] # print height_map print 'height std deviation:',math.sqrt(height_map[np.where(height_map>0.)].var()) # connect_structure = np.empty((3,3),dtype='int') # connect_structure[:,:] = 1 # for i in range(gr.grid_shape[2,0]): # slice = gr.grid[:,:,i] # slice_filled = ni.binary_fill_holes(slice) # slice_edge = slice_filled - ni.binary_erosion(slice_filled,connect_structure) # gr.grid[:,:,i] = slice_edge*(i+1) # # height_map = gr.grid.max(2) * gr.resolution[2,0] # # print height_map # print 'contoured height std deviation:',math.sqrt(height_map[np.where(height_map>0.)].var()) # high_pts_2d = obj[0:2,np.where(obj[2,:]>max_height-0.005)[1].A1] #high_pts_1d = min_evec.T * high_pts_2d high_pts_1d = ut.norm(high_pts_2d) idx1 = np.argmin(high_pts_1d) pt1 = high_pts_2d[:,idx1] idx2 = np.argmax(high_pts_1d) pt2 = high_pts_2d[:,idx2] if np.linalg.norm(pt1)<np.linalg.norm(pt2): grasp_point = np.row_stack((pt1,np.matrix([max_height]))) else: grasp_point = np.row_stack((pt2,np.matrix([max_height]))) vec = centroid_2d-grasp_point[0:2,0] gripper_angle = -math.atan2(-vec[0,0],vec[1,0]) grasp_vec = vec/np.linalg.norm(vec) if display_list != None: pt1 = np.row_stack((pt1,np.matrix([max_height]))) pt2 = np.row_stack((pt2,np.matrix([max_height]))) # display_list.insert(0,pu.CubeCloud(pt1,(0,0,200),size=(0.005,0.005,0.005))) # display_list.insert(0,pu.CubeCloud(pt2,(200,0,200),size=(0.005,0.005,0.005))) if display_list != None: pts = gr.grid_to_points() size = resolution # size = resolution*2 # size[2,0] = size[2,0]*2 #display_list.insert(0,pu.PointCloud(pts,(200,0,0))) display_list.append(pu.CubeCloud(pts,color=(200,0,0),size=size)) display_list.append(pu.CubeCloud(grasp_point,(0,200,200),size=(0.007,0.007,0.007))) v = np.row_stack((grasp_vec,np.matrix([0.]))) min_end_pt1 = grasp_point + v*min_size/2. min_end_pt2 = grasp_point - v*min_size/2. max_evec = np.matrix((min_evec[1,0],-min_evec[0,0])).T pts_1d = max_evec.T * pts_2d max_size = pts_1d.max() - pts_1d.min() display_list.append(pu.Line(min_end_pt1,min_end_pt2,color=(0,255,0))) return grasp_point,gripper_angle #---------------------- testing functions ------------------- def test_vertical_plane_finding(): display_list = [] rot_angle = dict['rot_angle'] gr = create_vertical_plane_grid(pt,pos_list,scan_list,l1,l2,rotation_angle=rot_angle, display_list=display_list) vertical_plane_points(gr) plane_cloud = pu.PointCloud(gr.grid_to_points(),color=(0,150,0)) display_list.insert(0,plane_cloud) po3d.run(display_list) def test_find_door_handle(): display_list = [] rot_angle = dict['rot_angle'] # pt[2,0] += 0.15 print 'pt:',pt.A1.tolist() gr = create_vertical_plane_grid(pt,pos_list,scan_list,l1,l2,rotation_angle=rot_angle, display_list=display_list) grid_pts_cloud = pu.PointCloud(gr.grid_to_points(),(0,0,255)) display_list.insert(0,grid_pts_cloud) copy_gr = copy.deepcopy(gr) obj_pts_list = [] print 'pt:',pt.A1.tolist() # Do I want to change the occupancy threshold when I'm closer? (see the old function test_find_door_handle_close) handle_object = find_door_handle(gr,pt,obj_pts_list,rotation_angle=rot_angle, occupancy_threshold=1,resolution=np.matrix([0.02,0.0025,0.02]).T) copy_gr.remove_vertical_plane() stickout_pts_cloud = pu.PointCloud(copy_gr.grid_to_points(),(100,100,100)) display_list.insert(0,stickout_pts_cloud) for i,obj_pts in enumerate(obj_pts_list): print 'mean:', obj_pts.mean(1).A1.tolist() size = [0.02,0.0025,0.02] # look at param for find_door_handle # size=gr.resolution.A1.tolist() size[0] = size[0]*2 size[1] = size[1]*2 # size[2] = size[2]*2 display_list.append(pu.CubeCloud(obj_pts,color=color_list[i%len(color_list)],size=size)) laser_point_cloud = pu.CubeCloud(pt,color=(0,200,0),size=(0.005,0.005,0.005)) po3d.run(display_list) #po3d.save(display_list, raw_name+'.png') def test_segmentation(): gr = create_segmentation_grid(pt,pos_list,scan_list,l1,l2, display_flag=True) obj_pts_list = segment_objects_points(gr) if obj_pts_list == None: print 'There is no plane' obj_pts_list = [] pts = gr.grid_to_points() d3m.plot_points(pts,color=(1.,1.,1.)) d3m.plot_points(pt,color=(0,1,0.),mode='sphere') for i,obj_pts in enumerate(obj_pts_list): size=gr.resolution.A1.tolist() size[2] = size[2]*2 d3m.plot_points(obj_pts,color=color_list[i%len(color_list)]) # display_list.append(pu.CubeCloud(obj_pts,color=color_list[i%len(color_list)],size=size)) #display_list.insert(0,pu.PointCloud(obj_pts,color=color_list[i%len(color_list)])) print 'mean:', obj_pts.mean(1).T d3m.show() def test_grasp_location_on_object(): display_list = [] # display_list = None gr = create_segmentation_grid(pt,pos_list,scan_list,l1,l2,display_list=display_list) obj_pts_list = segment_objects_points(gr) closest_obj = find_closest_object(obj_pts_list,pt) grasp_location_on_object(closest_obj,display_list) po3d.run(display_list) def test_plane_finding(): ''' visualize plane finding. ''' # brf = pt + np.matrix([-0.4,-0.2,-0.3]).T # brf[0,0] = max(brf[0,0],0.05) # print 'brf:', brf.T # # tlb = pt + np.matrix([0.3, 0.2,0.3]).T # resolution = np.matrix([0.01,0.01,0.0025]).T brf = pt+np.matrix([-0.15,-0.25,-0.2]).T brf[0,0] = max(0.07,brf[0,0]) tlb = pt+np.matrix([0.25, 0.25,0.2]).T resolution = np.matrix([0.01,0.01,0.0025]).T max_dist = np.linalg.norm(tlb) + 0.2 min_dist = brf[0,0] all_pts = generate_pointcloud(pos_list, scan_list, min_angle, max_angle, l1, l2,save_scan=False, max_dist=max_dist,min_dist=min_dist) #max_dist=2.0,min_dist=min_dist) gr = og3d.occupancy_grid_3d(brf,tlb,resolution) gr.fill_grid(all_pts) gr.to_binary(1) l = gr.find_plane_indices(assume_plane=True) z_min = min(l)*gr.resolution[2,0]+gr.brf[2,0] z_max = max(l)*gr.resolution[2,0]+gr.brf[2,0] print 'height of plane:', (z_max+z_min)/2 pts = gr.grid_to_points() plane_pts_bool = np.multiply(pts[2,:]>=z_min,pts[2,:]<=z_max) plane_pts = pts[:,np.where(plane_pts_bool)[1].A1.tolist()] above_pts =pts[:,np.where(pts[2,:]>z_max)[1].A1.tolist()] below_pts =pts[:,np.where(pts[2,:]<z_min)[1].A1.tolist()] d3m.plot_points(pt,color=(0,1,0.),mode='sphere') d3m.plot_points(plane_pts,color=(0,0,1.)) d3m.plot_points(above_pts,color=(1.0,1.0,1.0)) d3m.plot_points(below_pts,color=(1.,0.,0.)) cube_tups = gr.grid_lines() d3m.plot_cuboid(cube_tups) d3m.show() def test_approach(): display_list=[] # gr = create_approach_grid(pt,pos_list,scan_list,l1,l2,display_list=display_list,show_pts=False) gr = create_approach_grid(pt,pos_list,scan_list,l1,l2,display_list=display_list,show_pts=True) t0 = time.time() p_erratic,p_edge,h = find_goto_point_surface_1(gr,pt,display_list) t1 = time.time() print 'aaaaaaaaaaaaaah:', t1-t0 l = gr.find_plane_indices(assume_plane=True) z_min = min(l)*gr.resolution[2,0]+gr.brf[2,0] z_max = max(l)*gr.resolution[2,0]+gr.brf[2,0] print 'height of plane:', (z_max+z_min)/2 print 'height of surface in thok0 coord frame:', h print 'p_erratic in thok0:', p_erratic.T display_list.append(pu.CubeCloud(pt,color=(0,255,0),size=(0.018,0.018,0.018))) # display_list.append(pu.CubeCloud(p_erratic,color=(0,250,250),size=(0.007,0.007,0.007))) display_list.insert(0,pu.PointCloud(gr.grid_to_points(),color=(100,100,100))) po3d.run(display_list) def test_max_forward(): max_dist = math.sqrt(pt[0,0]**2+pt[1,0]**2+2.0**1) + 0.3 # max_dist = np.linalg.norm(pt[0:2]+0.3) min_angle,max_angle=math.radians(-40),math.radians(40) all_pts = generate_pointcloud(pos_list, scan_list, min_angle, max_angle, l1, l2, max_dist=max_dist, min_tilt=math.radians(-90)) display_list = [] max_x = max_fwd_without_collision(all_pts,0.20,max_dist,display_list) po3d.run(display_list) # print 'height_mat:', height_mat print 'max_x:', max_x dict = {'pos_list':pos_list, 'scan_list':scan_list,'l1':l1, 'l2':l2, 'pt':pt} # ut.save_pickle(dict,ut.formatted_time()+'_fwd_dict.pkl') def test_choose_grasp_strategy(): display_list = [] far_dist = 0.15 pt[1,0] += 0.0 gr = create_overhead_grasp_choice_grid(pt,pos_list,scan_list,l1,l2,far_dist,display_list) print 'overhead collide?',overhead_grasp_collision(pt,gr) display_list.append(pu.CubeCloud(pt,color=(0,200,0),size=(0.005,0.005,0.005))) # pts = generate_pointcloud(pos_list, scan_list, min_angle, max_angle, l1, l2,save_scan=False, # #max_dist=max_dist,min_dist=min_dist) # max_dist=2.0,min_dist=0.1) # display_list.append(pu.PointCloud(pts,color=(100,100,100))) po3d.run(display_list) def test_different_surfaces(): display_list = [] # all_pts = generate_pointcloud(pos_list, scan_list, math.radians(-40), math.radians(40), l1, l2,save_scan=False, # max_dist=np.Inf, min_dist=-np.Inf,min_tilt=-np.Inf,max_tilt=np.Inf) ## pts = all_pts[:,np.where(np.multiply(all_pts[0,:]>0.1,all_pts[0,:]<0.4))[1].A1] ## pts = pts[:,np.where(np.multiply(pts[1,:]<0.3,pts[1,:]>-0.3))[1].A1] ## pts = pts[:,np.where(pts[2,:]>-0.2)[1].A1] ## display_list.append(pu.PointCloud(pts,color=(0,200,0))) # display_list.append(pu.PointCloud(all_pts,color=(200,0,0))) # # brf = np.matrix([0.05,-0.35,-0.2]).T # tlb = np.matrix([0.5, 0.35,0.0]).T # resolution = np.matrix([0.01,0.01,0.0025]).T # gr = og3d.occupancy_grid_3d(brf,tlb,resolution) # gr.fill_grid(all_pts) # gr.to_binary(1) # gr = create_segmentation_grid(pt,pos_list,scan_list,l1,l2,display_list) gr = create_approach_grid(pt,pos_list,scan_list,l1,l2,display_list) l = gr.find_plane_indices(assume_plane=True) max_index = min(max(l)+5,gr.grid_shape[2,0]-1) min_index = max(min(l)-5,0) l = range(min_index,max_index+1) n_points_list = [] height_list = [] for idx in l: n_points_list.append(gr.grid[:,:,idx].sum()) height_list.append(idx*gr.resolution[2,0]+gr.brf[2,0]) pl.bar(height_list,n_points_list,width=gr.resolution[2,0],linewidth=0,align='center',color='y') max_occ = max(n_points_list) thresh = max_occ/5 xmin,xmax = pl.xlim() t = pl.axis() t = (xmin+0.0017,xmax-0.001,t[2],t[3]+50) # pl.plot([height_list[0],height_list[-1]],[thresh,thresh],c='r') pl.plot([xmin,xmax],[thresh,thresh],c='b') pl.title('Histogram of number of points vs z-coordinate of points') pl.xlabel('z-coordinate (relative to the laser range finder) (meters)') pl.ylabel('Number of points') pl.axis(t) pl.savefig(pkl_file_name+'.png') # pl.show() # print 'Mean:', pts.mean(1).T # pts_zeromean = pts-pts.mean(1) # n_points = pts.shape[1] # print 'n_points:', n_points # e_vals,e_vecs = np.linalg.eig(pts_zeromean*pts_zeromean.T/n_points) # # min_index = np.argmin(e_vals) # min_evec = e_vecs[:,min_index] # print 'min eigenvector:', min_evec.T # print 'min eigenvalue:', e_vals[min_index] # pts_1d = min_evec.T * pts # size = pts_1d.max() - pts_1d.min() # print 'spread along min eigenvector:', size # po3d.run(display_list) def test_occ_grid(): gr = create_approach_grid(pt,pos_list,scan_list,l1,l2,display_list=None,show_pts=True) pts = gr.grid_to_points() display_list=[] display_list.append(pu.CubeCloud(pt,color=(0,255,0),size=(0.007,0.007,0.007))) display_list.append(pu.PointCloud(pts,color=(200,0,0))) po3d.run(display_list) if __name__ == '__main__': p = optparse.OptionParser() p.add_option('-f', action='store', type='string', dest='pkl_file_name', help='file.pkl File with the scan,pos dict.') p.add_option('--all_pts', action='store_true', dest='show_all_pts', help='show all the points in light grey') opt, args = p.parse_args() pkl_file_name = opt.pkl_file_name show_full_cloud = opt.show_all_pts str_parts = pkl_file_name.split('.') raw_name = str_parts[-2] str_parts = raw_name.split('/') raw_name = str_parts[-1] dict = ut.load_pickle(pkl_file_name) pos_list = dict['pos_list'] scan_list = dict['scan_list'] min_angle = math.radians(-40) max_angle = math.radians(40) l1 = dict['l1'] l2 = dict['l2'] # l2 = -0.055 # l2 = 0.035 if dict.has_key('pt'): pt = dict['pt'] print 'dict has key pt' else: print 'dict does NOT have key pt' pt = np.matrix([0.35,0.0,-0.3]).T dict['pt'] = pt ut.save_pickle(dict,pkl_file_name) # charlie_nih() # test_grasp_location_on_object() # test_find_door_handle() # test_vertical_plane_finding_close() # test_vertical_plane_finding() test_segmentation() # test_plane_finding() # test_max_forward() # test_approach() # test_choose_grasp_strategy() # test_different_surfaces() # test_occ_grid()
[ [ 1, 0, 0.0261, 0.0009, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0261, 0.0009, 0, 0.66, 0.0227, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.027, 0.0009, 0, 0.6...
[ "import roslib; roslib.load_manifest('hrl_tilting_hokuyo')", "import roslib; roslib.load_manifest('hrl_tilting_hokuyo')", "import hrl_hokuyo.hokuyo_processing as hp", "import sys, optparse, os", "import hrl_lib.util as ut", "import hrl_lib.transforms as tr", "import numpy as np,math", "import time", ...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import sys, optparse, os import time import math, numpy as np import scipy.ndimage as ni import copy import hrl_lib.util as ut, hrl_lib.transforms as tr ## subtract occupancy grids. og1 = og1-og2 # # @param og1 - occupancy_grid_3d object. # @param og2 - occupancy_grid_3d object. # #will position og2 at an appropriate location within og1 (hopefully) #will copy points in og2 but not in og1 into og1 # # points corresponding to the gird cells whose occupancy drops to # zero will still be in grid_points_list #UNTESTED: # * subtracting grids of different sizes. # * how the rotation_z of the occupancy grid will affect things. def subtract(og1,og2): if np.all(og1.resolution==og2.resolution) == False: print 'occupancy_grid_3d.subtract: The resolution of the two grids is not the same.' print 'res1, res2:', og1.resolution.A1.tolist(), og2.resolution.A1.tolist() print 'Exiting...' sys.exit() sub_tlb = og2.tlb sub_brf = og2.brf sub_tlb_idx = np.round((sub_tlb-og1.brf)/og1.resolution) sub_brf_idx = np.round((sub_brf-og1.brf)/og1.resolution) x_s,y_s,z_s = int(sub_brf_idx[0,0]),int(sub_brf_idx[1,0]),int(sub_brf_idx[2,0]) x_e,y_e,z_e = int(sub_tlb_idx[0,0]),int(sub_tlb_idx[1,0]),int(sub_tlb_idx[2,0]) x_e = min(x_e+1,og1.grid_shape[0,0]) y_e = min(y_e+1,og1.grid_shape[1,0]) z_e = min(z_e+1,og1.grid_shape[2,0]) sub_grid = og1.grid[x_s:x_e,y_s:y_e,z_s:z_e] if np.any(og1.grid_shape!=og2.grid_shape): print '#############################################################################' print 'WARNING: occupancy_grid_3d.subtract has not been tested for grids of different sizes.' print '#############################################################################' sub_grid = sub_grid-og2.grid sub_grid = np.abs(sub_grid) # for now. og1.grid[x_s:x_e,y_s:y_e,z_s:z_e] = sub_grid idxs = np.where(sub_grid>=1) shp = og2.grid_shape list_idxs = (idxs[0]+idxs[1]*shp[0,0]+idxs[2]*shp[0,0]*shp[1,0]).tolist() og1_list_idxs = (idxs[0]+x_s+(idxs[1]+y_s)*shp[0,0]+(idxs[2]+z_s)*shp[0,0]*shp[1,0]).tolist() og1_list_len = len(og1.grid_points_list) for og1_pts_idxs,pts_idxs in zip(og1_list_idxs,list_idxs): if og1_pts_idxs<og1_list_len: og1.grid_points_list[og1_pts_idxs] += og2.grid_points_list[pts_idxs] ## class which implements the occupancy grid class occupancy_grid_3d(): ## # @param brf - 3x1 matrix. Bottom Right Front. # @param tlb - 3x1 matrix (coord of center of the top left back cell) # @param resolution - 3x1 matrix. size of each cell (in meters) along # the different directions. def __init__(self, brf, tlb, resolution, rotation_z=math.radians(0.)): #print np.round((tlb-brf)/resolution).astype('int')+1 self.grid = np.zeros(np.round((tlb-brf)/resolution).astype('int')+1,dtype='int') self.tlb = tlb self.brf = brf self.grid_shape = np.matrix(self.grid.shape).T self.resolution = resolution n_cells = self.grid.shape[0]*self.grid.shape[1]*self.grid.shape[2] self.grid_points_list = [[] for i in range(n_cells)] self.rotation_z = rotation_z ## returns list of 8 tuples of 3x1 points which form the edges of the grid. # Useful for displaying the extents of the volume of interest (VOI). # @return list of 8 tuples of 3x1 points which form the edges of the grid. def grid_lines(self, rotation_angle=0.): grid_size = np.multiply(self.grid_shape,self.resolution) rot_mat = tr.rotZ(rotation_angle) p5 = self.tlb p6 = p5+np.matrix([0.,-grid_size[1,0],0.]).T p8 = p5+np.matrix([0.,0.,-grid_size[2,0]]).T p7 = p8+np.matrix([0.,-grid_size[1,0],0.]).T p3 = self.brf p4 = p3+np.matrix([0.,grid_size[1,0],0.]).T p2 = p3+np.matrix([0.,0.,grid_size[2,0]]).T p1 = p2+np.matrix([0.,grid_size[1,0],0.]).T p1 = rot_mat*p1 p2 = rot_mat*p2 p3 = rot_mat*p3 p4 = rot_mat*p4 p5 = rot_mat*p5 p6 = rot_mat*p6 p7 = rot_mat*p7 p8 = rot_mat*p8 l = [(p1,p2),(p1,p4),(p2,p3),(p3,p4),(p5,p6),(p6,p7),(p7,p8),(p8,p5),(p1,p5),(p2,p6),(p4,p8),(p3,p7)] #l = [(p5,p6),(p5,p3),(p1,p2)] return l ## fill the occupancy grid. # @param pts - 3xN matrix of points. # @param ignore_z - not use the z coord of the points. grid will be like a 2D grid. # #each cell of the grid gets filled the number of points that fall in the cell. def fill_grid(self,pts,ignore_z=False): if ignore_z: idx = np.where(np.min(np.multiply(pts[0:2,:]>self.brf[0:2,:], pts[0:2,:]<self.tlb[0:2,:]),0))[1] else: idx = np.where(np.min(np.multiply(pts[0:3,:]>self.brf,pts[0:3,:]<self.tlb),0))[1] if idx.shape[1] == 0: print 'aha!' return pts = pts[:,idx.A1.tolist()] # Find coordinates p_all = np.round((pts[0:3,:]-self.brf)/self.resolution) # Rotate points pts[0:3,:] = tr.Rz(self.rotation_z).T*pts[0:3,:] for i,p in enumerate(p_all.astype('int').T): if ignore_z: p[0,2] = 0 if np.any(p<0) or np.any(p>=self.grid_shape.T): continue tup = tuple(p.A1) self.grid_points_list[ tup[0] + self.grid_shape[0,0] * tup[1] + self.grid_shape[0,0] * self.grid_shape[1,0] * tup[2]].append(pts[:,i]) self.grid[tuple(p.A1)] += 1 def to_binary(self,thresh=1): ''' all cells with occupancy>=thresh set to 1, others set to 0. ''' filled = (self.grid>=thresh) self.grid[np.where(filled==True)] = 1 self.grid[np.where(filled==False)] = 0 def argmax_z(self,index_min=-np.Inf,index_max=np.Inf,search_up=False,search_down=False): ''' searches in the z direction for maximum number of cells with occupancy==1 call this function after calling to_binary() returns index. ''' index_min = int(max(index_min,0)) index_max = int(min(index_max,self.grid_shape[2,0]-1)) z_count_mat = [] #for i in xrange(self.grid_shape[2,0]): for i in xrange(index_min,index_max+1): z_count_mat.append(np.where(self.grid[:,:,i]==1)[0].shape[0]) if z_count_mat == []: return None z_count_mat = np.matrix(z_count_mat).T max_z = np.argmax(z_count_mat) max_count = z_count_mat[max_z,0] max_z += index_min print '#### max_count:', max_count if search_up: max_z_temp = max_z for i in range(1,5): #if (z_count_mat[max_z+i,0]*3.0)>max_count: #A #if (z_count_mat[max_z+i,0]*8.0)>max_count: #B if (max_z+i)>index_max: break if (z_count_mat[max_z+i-index_min,0]*5.0)>max_count: #B' max_z_temp = max_z+i max_z = max_z_temp if search_down: max_z_temp = max_z for i in range(1,5): if (max_z-i)<index_min: break if (max_z-i)>index_max: continue if (z_count_mat[max_z-i-index_min,0]*5.0)>max_count: max_z_temp = max_z-i max_z = max_z_temp return max_z,max_count def find_plane_indices(self,hmin=-np.Inf,hmax=np.Inf,assume_plane=False): ''' assume_plane - always return something. returns list of indices (z) corrresponding to horizontal plane points. returns [] if there is no plane ''' index_min = int(max(round((hmin-self.brf[2,0])/self.resolution[2,0]),0)) index_max = int(min(round((hmax-self.brf[2,0])/self.resolution[2,0]),self.grid_shape[2,0]-1)) z_plane,max_count = self.argmax_z(index_min,index_max,search_up=True) if z_plane == None: print 'oink oink.' return [] #---------- A # extra_remove_meters = 0.01 # n_more_to_remove = int(round(extra_remove_meters/self.resolution[2,0])) # l = range(max(z_plane-n_more_to_remove-1,0), # min(z_plane+n_more_to_remove+1,self.grid_shape[2,0]-1)) #---------- B extra_remove_meters = 0.005 n_more_to_remove = int(round(extra_remove_meters/self.resolution[2,0])) l = range(max(z_plane-10,0), min(z_plane+n_more_to_remove+1,self.grid_shape[2,0]-1)) # figure out whether this is indeed a plane. if assume_plane == False: n_more = int(round(0.1/self.resolution[2,0])) l_confirm = l+ range(max(l),min(z_plane+n_more+1,self.grid_shape[2,0]-1)) grid_2d = np.max(self.grid[:,:,l],2) n_plane_cells = grid_2d.sum() grid_2d = ni.binary_fill_holes(grid_2d) # I want 4-connectivity while filling holes. n_plane_cells = grid_2d.sum() min_plane_pts_threshold = (self.grid_shape[0,0]*self.grid_shape[1,0])/4 print '###n_plane_cells:', n_plane_cells print 'min_plane_pts_threshold:', min_plane_pts_threshold print 'find_plane_indices grid shape:',self.grid_shape.T if n_plane_cells < min_plane_pts_threshold: print 'occupancy_grid_3d.find_plane_indices: There is no plane.' print 'n_plane_cells:', n_plane_cells print 'min_plane_pts_threshold:', min_plane_pts_threshold l = [] return l ## get centroids of all the occupied cells as a 3xN np matrix # @param occupancy_threshold - number of points in a cell for it to be "occupied" # @return 3xN matrix of 3d coord of the cells which have occupancy >= occupancy_threshold def grid_to_centroids(self,occupancy_threshold=1): p = np.matrix(np.row_stack(np.where(self.grid>=occupancy_threshold))).astype('float') p[0,:] = p[0,:]*self.resolution[0,0] p[1,:] = p[1,:]*self.resolution[1,0] p[2,:] = p[2,:]*self.resolution[2,0] p += self.brf return p def grid_to_points(self,array=None,occupancy_threshold=1): ''' array - if not None then this will be used instead of self.grid returns 3xN matrix of 3d coord of the cells which have occupancy >= occupancy_threshold ''' if array == None: array = self.grid idxs = np.where(array>=occupancy_threshold) list_idxs = (idxs[0]+idxs[1]*self.grid_shape[0,0]+idxs[2]*self.grid_shape[0,0]*self.grid_shape[1,0]).tolist() l = [] for pts_idxs in list_idxs: l += self.grid_points_list[pts_idxs] if l == []: p = np.matrix([]) else: p = np.column_stack(l) return p def labeled_array_to_points(self,array,label): ''' returns coordinates of centers of grid cells corresponding to label as a 3xN matrix. ''' idxs = np.where(array==label) list_idxs = (idxs[0]+idxs[1]*self.grid_shape[0,0]+idxs[2]*self.grid_shape[0,0]*self.grid_shape[1,0]).tolist() l = [] for pts_idxs in list_idxs: l += self.grid_points_list[pts_idxs] if l == []: p = np.matrix([]) else: p = np.column_stack(l) return p def remove_vertical_plane(self): ''' removes plane parallel to the YZ plane. changes grid. returns plane_indices, slice corresponding to the vertical plane. points behind the plane are lost for ever! ''' self.grid = self.grid.swapaxes(2,0) self.grid_shape = np.matrix(self.grid.shape).T # z_max_first,max_count = self.argmax_z(search_up=False) # z_max_second,max_count_second = self.argmax_z(index_min=z_max_first+int(round(0.03/self.resolution[0,0])) ,search_up=False) z_max_first,max_count = self.argmax_z(search_down=False) z_max_second,max_count_second = self.argmax_z(index_min=z_max_first+int(round(0.035/self.resolution[0,0])) ,search_down=False) z_max_first,max_count = self.argmax_z(search_down=False) #z_max = self.argmax_z(search_up=True) if (max_count_second*1./max_count) > 0.3: z_max = z_max_second else: z_max = z_max_first print 'z_max_first', z_max_first print 'z_max_second', z_max_second print 'z_max', z_max more = int(round(0.03/self.resolution[0,0])) plane_indices = range(max(0,z_max-more),min(z_max+more,self.grid_shape[2,0])) self.grid = self.grid.swapaxes(2,0) self.grid_shape = np.matrix(self.grid.shape).T ver_plane_slice = self.grid[plane_indices,:,:] self.grid[plane_indices,:,:] = 0 max_x = max(plane_indices) behind_indices = range(max_x,self.grid_shape[0,0]) self.grid[behind_indices,:,:] = 0 return plane_indices,ver_plane_slice def remove_horizontal_plane(self, remove_below=True,hmin=-np.Inf,hmax=np.Inf, extra_layers=0): ''' call after to_binary() removes points corresponding to the horizontal plane from the grid. remove_below - remove points below the plane also. hmin,hmax - min and max possible height of the plane. (meters) This function changes grid. extra_layers - number of layers above the plane to remove. Sometimes I want to be over zealous while removing plane points. e.g. max_fwd_without_collision it returns the slice which has been set to zero, in case you want to leave the grid unchanged. ''' l = self.find_plane_indices(hmin,hmax) if l == []: print 'occupancy_grid_3d.remove_horizontal_plane: No plane found.' return None,l add_num = min(10,self.grid_shape[2,0]-max(l)-1) max_l = max(l)+add_num l_edge = l+range(max(l),max_l+1) grid_2d = np.max(self.grid[:,:,l_edge],2) # grid_2d = ni.binary_dilation(grid_2d,iterations=1) # I want 4-connectivity while filling holes. grid_2d = ni.binary_fill_holes(grid_2d) # I want 4-connectivity while filling holes. connect_structure = np.empty((3,3),dtype='int') connect_structure[:,:] = 1 eroded_2d = ni.binary_erosion(grid_2d,connect_structure,iterations=2) grid_2d = grid_2d-eroded_2d idxs = np.where(grid_2d!=0) if max_l>max(l): for i in range(min(5,add_num)): self.grid[idxs[0],idxs[1],max(l)+i+1] = 0 if remove_below: l = range(0,min(l)+1)+l max_z = max(l) for i in range(extra_layers): l.append(max_z+i+1) l_edge = l+range(max(l),max_l+1) plane_and_below_pts = self.grid[:,:,l_edge] self.grid[:,:,l] = 0 # set occupancy to zero. return plane_and_below_pts,l_edge def segment_objects(self, twod=False): ''' segments out objects after removing the plane. call after calling to_binary. returns labelled_array,n_labels labelled_array - same dimen as occupancy grid, each object has a different label. ''' plane_and_below_pts,l = self.remove_horizontal_plane(extra_layers=0) if l == []: print 'occupancy_grid_3d.segment_objects: There is no plane.' return None,None if twod == False: labelled_arr,n_labels = self.find_objects() else: labelled_arr,n_labels = self.find_objects_2d() self.grid[:,:,l] = plane_and_below_pts return labelled_arr,n_labels def find_objects_2d(self): ''' projects all points into the xy plane and then performs segmentation by region growing. ''' connect_structure = np.empty((3,3),dtype='int') connect_structure[:,:] = 1 grid_2d = np.max(self.grid[:,:,:],2) # grid_2d = ni.binary_erosion(grid_2d) # grid_2d = ni.binary_erosion(grid_2d,connect_structure) labeled_arr,n_labels = ni.label(grid_2d,connect_structure) print 'found %d objects'%(n_labels) labeled_arr_3d = self.grid.swapaxes(2,0) labeled_arr_3d = labeled_arr_3d.swapaxes(1,2) print 'labeled_arr.shape:',labeled_arr.shape print 'labeled_arr_3d.shape:',labeled_arr_3d.shape labeled_arr_3d = labeled_arr_3d*labeled_arr labeled_arr_3d = labeled_arr_3d.swapaxes(2,0) labeled_arr_3d = labeled_arr_3d.swapaxes(1,0) labeled_arr = labeled_arr_3d # I still want to count cells in 3d (thin but tall objects.) if n_labels > 0: labels_list = range(1,n_labels+1) #count_objects = ni.sum(grid_2d,labeled_arr,labels_list) count_objects = ni.sum(self.grid,labeled_arr,labels_list) if n_labels == 1: count_objects = [count_objects] t0 = time.time() new_labels_list = [] for c,l in zip(count_objects,labels_list): if c > 3: new_labels_list.append(l) else: labeled_arr[np.where(labeled_arr == l)] = 0 # relabel stuff for nl,l in enumerate(new_labels_list): labeled_arr[np.where(labeled_arr == l)] = nl+1 n_labels = len(new_labels_list) t1 = time.time() print 'time:', t1-t0 print 'found %d objects'%(n_labels) # return labeled_arr,n_labels return labeled_arr_3d,n_labels def find_objects(self): ''' region growing kind of thing for segmentation. Useful if plane has been removed. ''' connect_structure = np.empty((3,3,3),dtype='int') grid = copy.copy(self.grid) connect_structure[:,:,:] = 0 connect_structure[1,1,:] = 1 iterations = int(round(0.005/self.resolution[2,0])) # iterations=5 #grid = ni.binary_closing(grid,connect_structure,iterations=iterations) connect_structure[:,:,:] = 1 labeled_arr,n_labels = ni.label(grid,connect_structure) print 'ho!' print 'found %d objects'%(n_labels) if n_labels == 0: return labeled_arr,n_labels labels_list = range(1,n_labels+1) count_objects = ni.sum(grid,labeled_arr,labels_list) if n_labels == 1: count_objects = [count_objects] # t0 = time.time() # remove_labels = np.where(np.matrix(count_objects) <= 5)[1].A1.tolist() # for r in remove_labels: # labeled_arr[np.where(labeled_arr == r)] = 0 # t1 = time.time() # labeled_arr,n_labels = ni.label(labeled_arr,connect_structure) # print 'time:', t1-t0 t0 = time.time() new_labels_list = [] for c,l in zip(count_objects,labels_list): if c > 3: new_labels_list.append(l) else: labeled_arr[np.where(labeled_arr == l)] = 0 # relabel stuff for nl,l in enumerate(new_labels_list): labeled_arr[np.where(labeled_arr == l)] = nl+1 n_labels = len(new_labels_list) t1 = time.time() print 'time:', t1-t0 print 'found %d objects'%(n_labels) return labeled_arr,n_labels if __name__ == '__main__': import pygame_opengl_3d_display as po3d import hokuyo.pygame_utils as pu import processing_3d as p3d p = optparse.OptionParser() p.add_option('-f', action='store', type='string', dest='pkl_file_name', help='file.pkl File with the scan,pos dict.',default=None) p.add_option('-c', action='store', type='string', dest='pts_pkl', help='pkl file with 3D points',default=None) opt, args = p.parse_args() pts_pkl = opt.pts_pkl pkl_file_name = opt.pkl_file_name #-------------- simple test --------------- # gr = occupancy_grid_3d(np.matrix([0.,0.,0]).T, np.matrix([1.,1.,1]).T, # np.matrix([1,1,1]).T) # pts = np.matrix([[1.1,0,-0.2],[0,0,0],[0.7,0.7,0.3],[0.6,0.8,-0.2]]).T # gr.fill_grid(pts) ## print gr.grid resolution = np.matrix([0.01,0.01,0.01]).T gr = occupancy_grid_3d(np.matrix([0.45,-0.5,-1.0]).T, np.matrix([0.65,0.05,-0.2]).T, resolution) if pts_pkl != None: pts = ut.load_pickle(pts_pkl) elif pkl_file_name != None: dict = ut.load_pickle(pkl_file_name) pos_list = dict['pos_list'] scan_list = dict['scan_list'] min_angle = math.radians(-40) max_angle = math.radians(40) l1 = dict['l1'] l2 = dict['l2'] pts = p3d.generate_pointcloud(pos_list, scan_list, min_angle, max_angle, l1, l2) else: print 'specify a pkl file -c or -f' print 'Exiting...' sys.exit() print 'started filling the grid' t0 = time.time() gr.fill_grid(pts) t1 = time.time() print 'time to fill the grid:', t1-t0 #grid_pts = gr.grid_to_points() grid_pts = gr.grid_to_centroids() ## print grid_pts cloud = pu.CubeCloud(grid_pts,(0,0,0),(resolution/2).A1.tolist()) pc = pu.PointCloud(pts,(100,100,100)) lc = pu.LineCloud(gr.grid_lines(),(100,100,0)) po3d.run([cloud,pc,lc])
[ [ 1, 0, 0.0489, 0.0016, 0, 0.66, 0, 509, 0, 3, 0, 0, 509, 0, 0 ], [ 1, 0, 0.0506, 0.0016, 0, 0.66, 0.125, 654, 0, 1, 0, 0, 654, 0, 0 ], [ 1, 0, 0.0538, 0.0016, 0, 0...
[ "import sys, optparse, os", "import time", "import math, numpy as np", "import scipy.ndimage as ni", "import copy", "import hrl_lib.util as ut, hrl_lib.transforms as tr", "def subtract(og1,og2):\n\n if np.all(og1.resolution==og2.resolution) == False:\n print('occupancy_grid_3d.subtract: The re...
# Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # ## author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('zenither') import rospy from hrl_msgs.msg import FloatArray from threading import RLock class ZenitherClient(): def __init__(self, init_ros_node = False, zenither_pose_topic = 'zenither_pose'): if init_ros_node: rospy.init_node('ZenitherClient') self.h = None self.lock = RLock() rospy.Subscriber(zenither_pose_topic, FloatArray, self.pose_cb) def pose_cb(self, fa): self.lock.acquire() self.h = fa.data[0] self.lock.release() def height(self): self.lock.acquire() h = self.h self.lock.release() return h if __name__ == '__main__': zc = ZenitherClient(init_ros_node = True) while not rospy.is_shutdown(): print 'h:', zc.height() rospy.sleep(0.1)
[ [ 1, 0, 0.4746, 0.0169, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.4746, 0.0169, 0, 0.66, 0.1667, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.4915, 0.0169, 0, 0....
[ "import roslib; roslib.load_manifest('zenither')", "import roslib; roslib.load_manifest('zenither')", "import rospy", "from hrl_msgs.msg import FloatArray", "from threading import RLock", "class ZenitherClient():\n def __init__(self, init_ros_node = False,\n zenither_pose_topic = 'zenit...
#!/usr/bin/python # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # ROS wrapper for zenither. ## author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) ## author Travis Deyle (Healthcare Robotics Lab, Georgia Tech.) from threading import RLock import roslib roslib.load_manifest('zenither') import zenither_config as zc import rospy from hrl_srvs.srv import Float_Int from hrl_msgs.msg import FloatArray class ZenitherClient(): def __init__(self, robot): try: rospy.init_node('ZenitherClient') rospy.logout('ZenitherServer: Initialized Node') except rospy.ROSException: pass if robot not in zc.calib: raise RuntimeError('unknown robot') self.calib = zc.calib[robot] srv = '/zenither/move_position' rospy.wait_for_service(srv) self.move_position = rospy.ServiceProxy(srv, Float_Int) srv = '/zenither/stop' rospy.wait_for_service(srv) self.stop = rospy.ServiceProxy(srv, Float_Int) srv = '/zenither/apply_torque' rospy.wait_for_service(srv) self.apply_torque = rospy.ServiceProxy(srv, Float_Int) srv = '/zenither/torque_move_position' rospy.wait_for_service(srv) self.torque_move_position = rospy.ServiceProxy(srv, Float_Int) zenither_pose_topic = 'zenither_pose' self.h = None self.lock = RLock() rospy.Subscriber(zenither_pose_topic, FloatArray, self.pose_cb) #---------- functions to send zenither commands. ------------- def estop(self): self.stop(0) def zenith(self, torque=None): if torque == None: torque=self.calib['zenith_torque'] self.apply_torque(torque) def nadir(self, torque=None): if torque == None: torque=self.calib['nadir_torque'] self.apply_torque(torque) #--------- zenither height functions -------------- def pose_cb(self, fa): self.lock.acquire() self.h = fa.data[0] self.lock.release() ## return the current height of the zenither. def height(self): self.lock.acquire() h = self.h self.lock.release() return h if __name__ == '__main__': import time cl = ZenitherClient('HRL2') cl.zenith() time.sleep(0.5) cl.estop()
[ [ 1, 0, 0.2845, 0.0086, 0, 0.66, 0, 83, 0, 1, 0, 0, 83, 0, 0 ], [ 1, 0, 0.3017, 0.0086, 0, 0.66, 0.125, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.3103, 0.0086, 0, 0.6...
[ "from threading import RLock", "import roslib", "roslib.load_manifest('zenither')", "import zenither_config as zc", "import rospy", "from hrl_srvs.srv import Float_Int", "from hrl_msgs.msg import FloatArray", "class ZenitherClient():\n def __init__(self, robot):\n try:\n rospy.ini...
calib = { 'El-E': { 'robot': 'El-E', 'pos_factor': 0.9144 / (183897 - 1250), 'vel_factor': 0.9144 / (183897 - 1250) / 20, 'acc_factor': 0.9144 / (183897 - 1250), 'POS_MAX': 0.9, 'VEL_DEFAULT': 1.5, 'VEL_MAX': 4.0, 'ACC_DEFAULT':0.0002, 'ACC_MAX':0.001, 'ZERO_BIAS':-0.004, 'HAS_BRAKE':True, 'nadir_torque': 0, 'zenith_torque': 200, 'down_fast_torque': 0, 'down_slow_torque': 50, 'down_snail_torque': 60, 'up_fast_torque': 300, 'up_slow_torque': 250, 'up_snail_torque': 200, 'max_height': 0.89, 'min_height': 0.005 }, 'HRL2': { #-------- Cressel's explanation ------------------------- # max speed for size 40 1m actuator is 1400rpm / 0.45m/s # conversion for velocity 536.87633 cnts/s / rpm # 536.87633*1400/0.45 # cnts/s / m/s (0.5,-540179)(0.0,0) # close but not there yet #-------- Advait's explanation # 1 rev of zenither = 20mm (from festo manual, page 10) # 1 rev of animatics servo = 2000 encoder counts (pg 6, section 1.0 of the animatics manual) # a gear reduction of 10 (I think I remember Cressel mentioning this). # => 20000 counts = 20mm or 1 count = 1/1000,000 meters # 'robot': 'HRL2', 'pos_factor': 1.0/ (-1000000), # Advait - Apr 27, 2009 'POS_MAX': 0.96, # Advait - Feb 17, 2009 'ZERO_BIAS':0.28, 'HAS_BRAKE':True, 'nadir_torque': 3, 'zenith_torque': 400, 'down_fast_torque': 0, 'down_slow_torque': 3, 'down_snail_torque': 5, 'up_fast_torque': 450, 'up_slow_torque': 400, 'up_snail_torque': 300, 'zero_vel_torque': 90, 'max_height': 1.32, 'min_height': 0.29 }, 'test_rig': { #-------- Advait's explanation # 1 rev of zenither = 10mm (from festo manual, page 10) # 1 rev of animatics servo = 2000 encoder counts (pg 6, section 1.0 of the animatics manual) # => 2000 counts = 10mm or 1 count = 1/200,000 meters # # for vel_factor and acc_factor, see Page 8 of Animatics manual # 'robot': 'test_rig', 'pos_factor': 1.0 /200000, 'vel_factor': 1.0/(32212.578*100), 'acc_factor': 1.0/(7.9166433*100), 'POS_MAX': 0.9, 'VEL_DEFAULT': 0.2, 'VEL_MAX': 0.4, 'ACC_DEFAULT':0.5, 'ACC_MAX':10.0, 'ZERO_BIAS':0.0, 'HAS_BRAKE':False, 'nadir_torque': -150, 'zenith_torque': 150, 'max_height': 0.9, 'min_height': 0.005 }, }
[ [ 14, 0, 0.5, 0.9643, 0, 0.66, 0, 147, 0, 0, 0, 0, 0, 6, 0 ] ]
[ "calib = {\n 'El-E': {\n 'robot': 'El-E',\n 'pos_factor': 0.9144 / (183897 - 1250),\n 'vel_factor': 0.9144 / (183897 - 1250) / 20,\n 'acc_factor': 0.9144 / (183897 - 1250),\n 'POS_MAX': 0.9,\n 'VEL_DEFAULT': 1.5," ]
__all__ = [ 'zenither', 'lib_zenither', 'ros_zenither_server', 'ros_zenither_client' ]
[ [ 14, 0, 0.5833, 1, 0, 0.66, 0, 272, 0, 0, 0, 0, 0, 5, 0 ] ]
[ "__all__ = [\n'zenither',\n'lib_zenither',\n'ros_zenither_server',\n'ros_zenither_client'\n]" ]
#!/usr/bin/python # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # ## author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) ## author Travis Deyle (Healthcare Robotics Lab, Georgia Tech.) ## author Hai Nguyen (Healthcare Robotics Lab, Georgia Tech.) import roslib roslib.load_manifest('zenither') import rospy from hrl_srvs.srv import Float_Int import time import sys class ZenitherServer(): def __init__(self, zenither): try: rospy.init_node('ZenitherServer') rospy.logout('ZenitherServer: Initialized Node') except rospy.ROSException: pass self.zenither = zenither self.smp = rospy.Service('/zenither/move_position', Float_Int, self.move_position) self.ss = rospy.Service('/zenither/stop', Float_Int, self.estop) self.sat = rospy.Service('/zenither/apply_torque', Float_Int, self.apply_torque) self.stmp = rospy.Service('/zenither/torque_move_position', Float_Int, self.torque_move_position) def move_position(self, msg): print 'move_position is UNTESTED' #self.zenither.move_position( msg.value ) return True def apply_torque(self, req): self.zenither.nadir(req.value) return True def estop(self, req): self.zenither.estop() return True def torque_move_position(self, req): self.zenither.torque_move_position(req.value) return True if __name__ == '__main__': import zenither.zenither as zenither z = zenither.Zenither('HRL2', pose_read_thread = True) zs = ZenitherServer(z) rospy.spin() # rosservice call /zenither/move_position 0.3
[ [ 1, 0, 0.3882, 0.0118, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.4, 0.0118, 0, 0.66, 0.1429, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.4118, 0.0118, 0, 0.66,...
[ "import roslib", "roslib.load_manifest('zenither')", "import rospy", "from hrl_srvs.srv import Float_Int", "import time", "import sys", "class ZenitherServer():\n def __init__(self, zenither):\n try:\n rospy.init_node('ZenitherServer')\n rospy.logout('ZenitherServer: Init...
#!/usr/bin/python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Cressel Anderson (Healthcare Robotics Lab, Georgia Tech.) # \author Marc Killpack (Healthcare Robotics Lab, Georgia Tech.) import pygame import pygame.joystick import sys, optparse import segway_command as sc import time as time import roslib roslib.load_manifest('segway_omni') import rospy import hrl_lib.util as ut import numpy as np from pygame.locals import * if __name__=='__main__': p = optparse.OptionParser() p.add_option('-z', action='store_true', dest='zenither', help='control the zenither also') opt, args = p.parse_args() zenither_flag = opt.zenither if zenither_flag: import zenither.zenither as zenither z = zenither.Zenither(robot='HRL2') cmd_node = sc.SegwayCommand() max_xvel = 0.18 max_yvel = 0.15 max_speed = 0.18 # don't exceed 0.18 under any condition. max_avel = 0.18 xvel = 0.0 yvel = 0.0 avel = 0.0 #init pygame pygame.init() #joystick_status joystick_count = pygame.joystick.get_count() print "Joysticks found: %d\n" % joystick_count try: js = pygame.joystick.Joystick(0) js.init() except pygame.error: print "joystick error" js = None js_status = js.get_init() print js_status screen = pygame.display.set_mode((320,80)) pygame.display.set_caption('Snozzjoy') background = pygame.Surface(screen.get_size()) background = background.convert() background.fill((250,250,250)) x=0. y=0. a=0. len = 5 xvel_history = np.matrix(np.zeros([len,1])) connected = False while not rospy.is_shutdown(): move_zenither_flag=False for event in pygame.event.get(): if (event.type == JOYAXISMOTION): if event.axis == 0: #left pedal x = -0.5 * event.value - 0.5 elif event.axis == 1 and x == 0: #right pedal x = 0.5 * event.value + 0.5 if event.axis == 2: a = event.value print 'event.axis: ',event.axis,'event.value: ',event.value # elif (event.type == JOYBUTTONDOWN): # if zenither_flag: # if(event.button == 0): # z.zenith(z.calib['up_slow_torque']) # move_zenither_flag=True # if(event.button == 1): # z.nadir(z.calib['down_snail_torque']) # move_zenither_flag=True if x == 0 and y == 0 and a ==0: connected = True # detect a joystick disconnect try: js.quit() js.init() except pygame.error: print "joystick error" rospy.signal_shutdown() # if zenither_flag and (move_zenither_flag == False): # z.estop() #send segway commands if connected: xvel = x*max_xvel # xvel_history[0:(len-1)] = xvel_history[1:len] # xvel_history[(len-1)] = xvel # xvel = np.mean(xvel_history) yvel = y*max_yvel avel = a*max_avel vel_vec = np.matrix([xvel,yvel]).T vel_mag = np.linalg.norm(vel_vec) speed = ut.bound(vel_mag,max_speed,0.) if speed >= 0.05: vel_vec = vel_vec*speed/vel_mag xvel,yvel = vel_vec[0,0],vel_vec[1,0] cmd_node.set_velocity(xvel,yvel,avel) print '*******xvel=',xvel,'; avel=',avel # stop the segway cmd_node.set_velocity(0.,0.,0.)
[ [ 1, 0, 0.1951, 0.0061, 0, 0.66, 0, 87, 0, 1, 0, 0, 87, 0, 0 ], [ 1, 0, 0.2012, 0.0061, 0, 0.66, 0.0909, 115, 0, 1, 0, 0, 115, 0, 0 ], [ 1, 0, 0.2073, 0.0061, 0, 0....
[ "import pygame", "import pygame.joystick", "import sys, optparse", "import segway_command as sc", "import time as time", "import roslib", "roslib.load_manifest('segway_omni')", "import rospy", "import hrl_lib.util as ut", "import numpy as np", "from pygame.locals import *", "if __name__=='__ma...
# Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Cressel Anderson (Healthcare Robotics Lab, Georgia Tech.) import usb import pickle import time import numpy as np import math import struct class Mecanum_Properties(): def __init__( self ): self.r1 = .2032/2. # 8in wheels self.r2 = .2032/2. # 8in wheels self.R = 4*2.54/100 # wheel radius in meters self.la = .6223/2 # 1/2 width of the base self.lb = .33655/2 # 1/2 length of the base class Mecanum( Mecanum_Properties ): def __init__( self ): self.segway_front = Segway( side='front') self.segway_back = Segway( side='back' ) Mecanum_Properties.__init__(self) self.get_status() def set_velocity( self, xvel, yvel, avel ): """xvel and yvel should be in m/s and avel should be rad/s""" yvel = -yvel R = self.R la = self.la lb = self.lb # eq. 17 from Knematic modeling for feedback control of an # omnidirectional wheeled mobile robot by Muir and Neuman at # CMU J = 1/R * np.matrix([-1,1, (la + lb), 1,1,-(la + lb), -1,1,-(la + lb), 1,1, (la + lb) ]).reshape(4,3) # their coordinate frame is rotated dP = np.matrix( [ -yvel, -xvel, avel] ).T w = J*dP # they use a goofy ordering flw = w[1,0] frw = w[0,0] blw = w[2,0] brw = w[3,0] front_cmd = self.segway_front.send_wheel_velocities( flw, frw ) back_cmd = self.segway_back.send_wheel_velocities( blw, brw ) if front_cmd == False or back_cmd == False: print 'Velocities out of spec: ',flw,frw,blw,brw print 'Perhaps we should consider sending the max command so motion doesn\'t hang.' def stop(self): self.set_velocity(0.,0.,0.) def get_status( self ): pass class Segway_Properties(): def __init__( self ): self._integrated_wheel_displacement = 24372/(2*math.pi) # countsp / rad #vv = 39 # countsp/sec / count_velocity --- original #tv = 19 # countsp/sec / count_velocity --- original vv = 39 # countsp/sec / count_velocity tv = 22 # countsp/sec / count_velocity self._V_vel = vv /self._integrated_wheel_displacement # rad/sec / count_velocity self._T_vel = tv /self._integrated_wheel_displacement # rad/sec / count_velocity # w = | a b | * V # | c d | a = self._V_vel b = self._T_vel c = self._V_vel d = -self._T_vel self.A = np.matrix([a,b,c,d]).reshape(2,2) self.A_inv = 1/(a*b-c*d) * np.matrix([ d, -b, -c, a ]).reshape(2,2) # w = A*V # V = A_inv*w # in addition to what should command the platform velocities should be self._power_base_battery_voltage = 1/4. #(volt)/count self._ui_battery_voltage = .0125 #(volt)/count self._ui_battery_voltage_offset = 1.4 #volts class Segway( Segway_Properties ): def __init__( self, side='front' ): """side should be 'front' or 'back'""" Segway_Properties.__init__(self) self.side = side self.segway = None self.connect() self.pitch_ang = 0 self.pitch_rate = 0 self.yaw_ang = 0 self.yaw_rate = 0 self.LW_vel = 0 self.RW_vel = 0 self.yaw_rate = 0 self.servo_frames = 0 self.integrated_wheel_displacement_left = 0 self.integrated_wheel_displacement_right = 0 self.integrated_for_aft_displacement = 0 self.integrated_yaw_displacement = 0 self.left_motor_torque = 0 self.right_motor_torque = 0 self.operational_mode = 0 self.controller_gain_schedule = 0 self.ui_battery_voltage = 0 self.power_base_battery_voltage = 0 self.velocity_commanded = 0 self.turn_commanded = 0 def connect(self): buses = usb.busses() segway_handle_list = [] for bus in buses: for device in bus.devices: if device.idVendor == 1027 and device.idProduct == 59177: h = device.open() serial_num = int(h.getString(3,10)) if serial_num == 215: if self.side == 'front': print 'Connected to front segway' self.segway = h self.segway.claimInterface(0) elif serial_num == 201: if self.side == 'back': print 'Connected to rear segway' self.segway = h self.segway.claimInterface(0) else: raise RuntimeError( 'Unknown_segway connected.' + ' Serial Number is ',serial_num ) def calc_checksum(self, msg): checksum = 0 for byt in msg: checksum = (checksum+byt)%65536 checksum_hi = checksum >> 8 checksum &= 0xff checksum = (checksum+checksum_hi)%65536 checksum_hi = checksum >> 8 checksum &= 0xff checksum = (checksum+checksum_hi)%65536 checksum = (~checksum+1)&0xff return checksum def compose_velocity_cmd(self,linvel,angvel): byt_hi = 0x04 byt_lo = 0x13 if self.side == 'back': # because the front segway is linvel = -linvel # turned around linvel_counts = int(linvel) angvel_counts = int(angvel) if abs(linvel_counts)>1176: print 'connect.compose_velocity_cmd: Linear velocity is too high. counts: %d'%linvel return [] if abs(angvel_counts)>1024: print 'connect.compose_velocity_cmd: Angular velocity is too high. counts: %d'%angvel return [] usb_msg_fixed = [0xf0,0x55,0x00,0x00,0x00,0x00,byt_hi,byt_lo,0x00] can_vel_msg = [(linvel_counts>>8)&0xff,linvel_counts&0xff,(angvel_counts>>8)&0xff,angvel_counts&0xff,0x00,0x00,0x00,0x00] msg_send = usb_msg_fixed + can_vel_msg msg_send.append(self.calc_checksum(msg_send)) return msg_send def send_command(self, linvel0, angvel0 ): msg0 = self.compose_velocity_cmd(linvel0,angvel0) if msg0 == []: return False for i in range(1): self.segway.bulkWrite(0x02,msg0) def send_wheel_velocities( self, lvel, rvel ): w = np.matrix([lvel,rvel]).T V = self.A_inv*w #print 'V = ',V xv = V[0,0] tv = V[1,0] #print 'Left vel = ',lvel #print 'Right vel = ',rvel #print 'Forward command = ',xv #print 'Turn command = ',tv return self.send_command( xv, tv ) def parse_usb_cmd(self,msg): if len(msg) < 18: return if self.calc_checksum(msg[:-1]) != msg[-1]: #got garbage rejecting return id = ((msg[4]<<3)|((msg[5]>>5)&7))&0xfff data = msg[9:17] if id == 0x400: # unused pass elif id == 0x401: self.pitch_ang = self._2_bytes( data[0], data[1] ) self.pitch_rate = self._2_bytes( data[2], data[3] ) self.yaw_ang = self._2_bytes( data[4], data[5] ) self.yaw_rate = self._2_bytes( data[6], data[7] ) elif id == 0x402: self.LW_vel = self._2_bytes( data[0], data[1] )#/self._LW_vel self.RW_vel = self._2_bytes( data[2], data[3] )#/self._RW_vel self.yaw_rate = self._2_bytes( data[4], data[5] ) self.servo_frames = self._2_bytes_unsigned( data[6], data[7] ) elif id == 0x403: self.integrated_wheel_displacement_left = self._4_bytes(data[2],data[3],data[0],data[1]) self.integrated_wheel_displacement_right = self._4_bytes(data[6],data[7],data[4],data[5]) pass elif id == 0x404: self.integrated_for_aft_displacement = self._4_bytes(data[2],data[3],data[0],data[1]) self.integrated_yaw_displacement = self._4_bytes(data[6],data[7],data[4],data[5]) elif id == 0x405: self.left_motor_torque = self._2_bytes( data[0], data[1] ) self.right_motor_torque = self._2_bytes( data[2], data[3] ) elif id == 0x406: self.operational_mode = self._2_bytes( data[0], data[1] ) self.controller_gain_schedule = self._2_bytes( data[2], data[3] ) self.ui_battery_voltage = self._2_bytes( data[4], data[5] )*self._ui_battery_voltage + self._ui_battery_voltage_offset self.power_base_battery_voltage = self._2_bytes( data[6], data[7] )*self._power_base_battery_voltage elif id == 0x407: self.velocity_commanded = self._2_bytes( data[0], data[1] )#/self._LW_vel self.turn_commanded = self._2_bytes( data[2], data[3] ) elif msg[1] == 0x00: # print 'heartbeat id = ',hex(msg[6]),hex(msg[7]) pass elif id == 0x680: # CU Status Message pass else: #print 'no message parsed:', hex(id) pass def _2_bytes( self,high, low ): return struct.unpack('>h',chr(high)+chr(low))[0] def _2_bytes_unsigned( self,high, low ): return struct.unpack('>H',chr(high)+chr(low))[0] def _4_bytes( self,highhigh, lowhigh, highlow, lowlow ): return struct.unpack('>l',chr(highhigh)+chr(lowhigh)+chr(highlow)+chr(lowlow))[0] def clear_read(self): rd = self.segway.bulkRead(0x81,1000) def read(self): before = time.time() rd = self.segway.bulkRead(0x81,9*(18+2)) msg = [(a & 0xff) for a in rd] i=0 while 1: try: msg.pop(i*62) msg.pop(i*62) i += 1 except IndexError: break #find the start idx1 = msg.index(0xf0) if msg[idx1+18] != 0xf0: # if this is not the start of a message get rid of the bad start msg = msg[idx1+1:] else: # we found the start while len(msg) > 17: try: single_msg = msg[idx1:idx1+18] if (single_msg[1] == 0x55 and single_msg[2] == 0xaa) or single_msg[1] == 0x00: self.parse_usb_cmd(single_msg) msg = msg[18:] except IndexError: break def print_feedback( self ): print 'self.integrated_wheel_displacement_left = ',self.integrated_wheel_displacement_left print 'self.integrated_wheel_displacement_right = ',self.integrated_wheel_displacement_right print 'self.LW_vel = ',self.LW_vel print 'self.RW_vel = ',self.RW_vel print 'frame = ',self.servo_frames print 'self.velocity_commanded = ',self.velocity_commanded print 'self.turn_commanded = ',self.turn_commanded #print 'self.yaw_rate = ',self.yaw_rate if __name__ == '__main__': import segway seg= segway.Segway() seg.clear_read() # send_command_fixed(segway_rear) rrates = [] lrates = [] for vel in range(31): linvel = 2000.*(vel)/30 - 1000. angvel = 0. start = time.time() last = start lastwrite = start lastread = start while 1: # seg.send_command(200,0.0) # seg.send_wheel_velocities(100.0,100.0) if time.time() - lastwrite > 0.1: print 'loop 1',time.time() - start #seg.send_command( linvel, angvel ) lastwrite = time.time() #seg.send_platform_velocities( 0,0 ) seg.send_wheel_velocities(1.,1.) if time.time() - lastread > 0.01: seg.read() lastread = time.time() #seg.print_feedback() if time.time() - start > 2.0: break left_start = seg.integrated_wheel_displacement_left right_start = seg.integrated_wheel_displacement_right first_points = time.time() while 1: # seg.send_command(200,0.0) # seg.send_wheel_velocities(100.0,100.0) if time.time() - lastwrite > 0.1: print 'loop 1.5',time.time() - start #seg.send_command( linvel, angvel ) lastwrite = time.time() #seg.send_platform_velocities( 0,0 ) seg.send_wheel_velocities(1.,1.) if time.time() - lastread > 0.01: seg.read() lastread = time.time() #seg.print_feedback() if time.time() - start > 5.0: break left_stop = seg.integrated_wheel_displacement_left right_stop = seg.integrated_wheel_displacement_right last_points = time.time() diff = last_points - first_points print 'Time: ',diff print 'left rate: ',(left_stop - left_start)/diff print 'right rate: ',(right_stop - right_start)/diff rrates.append(((right_stop - right_start)/diff,linvel)) lrates.append(((left_stop - left_start)/diff,linvel)) while 1: if time.time() - lastread > 0.01: seg.read() lastread = time.time() #seg.print_feedback() if seg.LW_vel ==0 and seg.RW_vel == 0: break print 'rrates:',rrates print 'lrates:',lrates import pylab x = [] y = [] x1 = [] y1 = [] for a, b in rrates: x.append(b) y.append(a) for a, b in lrates: x1.append(b) y1.append(a) pylab.plot(x,y,x1,y1) pylab.show() #while 1: # print 'loop 2' # seg.read() #seg.print_feedback() # print 'time: ', time.time()-start # if seg.LW_vel == 0 and seg.RW_vel == 0: # break # print 'total time: ', time.time()-start # print 'time to stop: ', time.time()-stop # msg = [] # while True: # msg += list(handle_rmp0.bulkRead(0x81,100)) # idx1 = msg.index(0xf0) # idx2 = msg.index(0xf0,idx1+1) # if idx2-idx1 == 18: # single_msg = msg[idx1:idx2] # if single_msg[1] == 0x55 and single_msg[2] == 0xaa: # parse_usb_cmd(single_msg) # # msg = msg[idx2:]
[ [ 1, 0, 0.0625, 0.0022, 0, 0.66, 0, 856, 0, 1, 0, 0, 856, 0, 0 ], [ 1, 0, 0.0647, 0.0022, 0, 0.66, 0.1, 848, 0, 1, 0, 0, 848, 0, 0 ], [ 1, 0, 0.0668, 0.0022, 0, 0.6...
[ "import usb", "import pickle", "import time", "import numpy as np", "import math", "import struct", "class Mecanum_Properties():\n def __init__( self ):\n self.r1 = .2032/2. # 8in wheels\n self.r2 = .2032/2. # 8in wheels\n \n self.R = 4*2.54/100 # wheel radius in meters\n ...
#! /usr/bin/python # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Hai Nguyen (Healthcare Robotics Lab, Georgia Tech.) # \author Marc Killpack (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('hrl_segway_omni') from hrl_lib.msg import PlanarBaseVel from geometry_msgs.msg import Twist import rospy import segway import numpy as np def callback(cmd): #print 'segway_node:', cmd.xvel, cmd.yvel, cmd.angular_velocity mecanum.set_velocity(cmd.xvel, cmd.yvel, cmd.angular_velocity) def callback_ros(cmd): #print 'segway_node:', cmd.linear.x, cmd.linear.y, cmd.angular.z avel = cmd.angular.z * 0.5 avel = np.clip(avel,-0.2,0.2) mecanum.set_velocity(cmd.linear.x, cmd.linear.y, avel) rospy.init_node("segway_node", anonymous=False) rospy.Subscriber("base", PlanarBaseVel, callback, None, 1) rospy.Subscriber("cmd_vel", Twist, callback_ros, None, 1) #mecanum = segway.Mecanum(init_ros_node=False) mecanum = segway.Mecanum() rospy.spin()
[ [ 1, 0, 0.566, 0.0189, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.566, 0.0189, 0, 0.66, 0.0769, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.5849, 0.0189, 0, 0.66...
[ "import roslib; roslib.load_manifest('hrl_segway_omni')", "import roslib; roslib.load_manifest('hrl_segway_omni')", "from hrl_lib.msg import PlanarBaseVel", "from geometry_msgs.msg import Twist", "import rospy", "import segway", "import numpy as np", "def callback(cmd):\n #print 'segway_node:', cmd...
#!/usr/bin/python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) # \author Cressel Anderson (Healthcare Robotics Lab, Georgia Tech.) # \author Hai Nguyen (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('force_torque') import rospy from hrl_msgs.msg import FloatArray #import hrl_lib.rutils as ru import AMTIForce2 as af import threading import time #import rutils as ru class AMTIForceServer(threading.Thread): def __init__(self, device_path, str_id): threading.Thread.__init__(self) try: rospy.init_node('AMTIForceServer') print 'AMTIForceServer: ros is up!' except rospy.ROSException: pass print 'AMTIForceServer: connecting to', device_path self.force_plate = af.AMTI_force(device_path) name = 'force_torque_' + str_id print 'AMTIForceServer: publishing', name, 'with type FloatArray' self.channel = rospy.Publisher(name, FloatArray, tcp_nodelay=True) def broadcast(self): print 'AMTIForceServer: started!' while not rospy.is_shutdown(): time.sleep(1/5000.0) self.channel.publish(FloatArray(None, self.force_plate.read().T.tolist()[0])) #DEPRECATED, use FTClient from ROSFTSensor with id = 0 #def AMTIForceClient(): # return ru.FloatArrayListener('AMTIForceClient', 'force_plate', 100.0) #import roslib; roslib.load_manifest('force_torque') #import force_torque.ROSAMTIForce as ft if __name__ == '__main__': import optparse p = optparse.OptionParser() p.add_option('--path', action='store', default='/dev/robot/force_plate0', type = 'string', dest='path', help='path to force torque device in (linux)') p.add_option('--name', action='store', default='ft1', type='string', dest='name', help='name given to FTSensor') opt, args = p.parse_args() server = AMTIForceServer(opt.path, opt.name) server.broadcast()
[ [ 1, 0, 0.3929, 0.0119, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.3929, 0.0119, 0, 0.66, 0.125, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.4048, 0.0119, 0, 0.6...
[ "import roslib; roslib.load_manifest('force_torque')", "import roslib; roslib.load_manifest('force_torque')", "import rospy", "from hrl_msgs.msg import FloatArray", "import AMTIForce2 as af", "import threading", "import time", "class AMTIForceServer(threading.Thread):\n def __init__(self, device_pa...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) # \author Cressel Anderson (Healthcare Robotics Lab, Georgia Tech.) # \author Hai Nguyen (Healthcare Robotics Lab, Georgia Tech.) import time, os, copy import serial, numpy as np import threading from threading import RLock class AMTI_force: def __init__(self, dev_name="/dev/robot/force_plate0", broadcast=True): self.dev_name = dev_name self.serial_dev = open_serial(dev_name, 57600) self.offset = None if not self._reset(): while not self._reset(): time.sleep(50/1000.0) self.last_value = self._read_once() def read(self): reading = self._read_once() while (reading == self.last_value).all(): time.sleep(8/1000.0) reading = self._read_once() return reading def _reset(self): self.serial_dev.write('R') self.serial_dev.flushInput() self.serial_dev.flushOutput() self.serial_dev.write('Q') x = self.serial_dev.read(1) print 'AMTI_force._reset: resetting.' return x == 'U' def _read_once(self): #Read until succeeds while True: x = self.serial_dev.read(24) if len(x) > 0: ret, error = stream_to_values(x) ret[0,0] = 4.448221628254617 * ret[0,0] ret[1,0] = 4.448221628254617 * ret[1,0] ret[2,0] = 4.448221628254617 * ret[2,0] ret[3,0] = 4.448221628254617 * 0.0254 * ret[3,0] ret[4,0] = 4.448221628254617 * 0.0254 * ret[4,0] ret[5,0] = 4.448221628254617 * 0.0254 * ret[5,0] if error: time.sleep(50/1000.0) self._reset() else: ft_vector = ret ft_vector[0,0] = -ft_vector[0,0] * 1.8 ft_vector[1,0] = ft_vector[1,0] * 1.7 ft_vector[2,0] = -ft_vector[2,0] ft_vector[3,0] = -ft_vector[3,0] ft_vector[5,0] = -ft_vector[5,0] return ft_vector COEFF_MAT = np.matrix([[ 0.0032206,-0.0000321,0.0003082,-0.0032960,-0.0000117,-0.0002678,0.0032446,0.0000940,0.0001793,-0.0031230,-0.0000715,-0.0002365], [ -0.0001470,0.0032134,0.0004389,0.0000222,0.0032134,0.0003946,0.0000684,-0.0031486,0.0000523,-0.0000209,-0.0031797,-0.0001470], [ -0.0006416,-0.0005812,-0.0087376,-0.0006207,-0.0005215,-0.0086779,-0.0006731,-0.0008607,-0.0087900,-0.0005766,-0.0007237,-0.0084300], [ 0.0000000,0.0000000,-0.0279669,0.0000000,0.0000000,-0.0269454,0.0000000,0.0000000,0.0281477,0.0000000,0.0000000,0.0268347], [ 0.0000000,0.0000000,0.0273877,0.0000000,0.0000000,-0.0269034,0.0000000,0.0000000,0.0278664,0.0000000,0.0000000,-0.0272117], [ -0.0123918,0.0124854,0.0000917,0.0126818,-0.0124854,0.0000911,0.0124843,-0.0122338,0.0000923,-0.0120165,0.0123544,0.0000885]]) def open_serial(dev_name, baudrate): serial_dev = serial.Serial(dev_name, timeout=4.) serial_dev.setBaudrate(baudrate) serial_dev.setParity('N') serial_dev.setStopbits(1) serial_dev.open() serial_dev.flushOutput() serial_dev.flushInput() serial_dev.write('R') time.sleep(1.) if(serial_dev == None): raise RuntimeError("AMTI_force: Serial port not found!\n") return serial_dev def stream_to_values(data): val = np.matrix(np.zeros((12,1))) val[0,0] = ord(data[0]) + 256*(ord(data[1])%16) val[1,0] = ord(data[2]) + 256*(ord(data[3])%16) val[2,0] = ord(data[4]) + 256*(ord(data[5])%16) val[3,0] = ord(data[6]) + 256*(ord(data[7])%16) val[4,0] = ord(data[8]) + 256*(ord(data[9])%16) val[5,0] = ord(data[10]) + 256*(ord(data[11])%16) val[6,0] = ord(data[12]) + 256*(ord(data[13])%16) val[7,0] = ord(data[14]) + 256*(ord(data[15])%16) val[8,0] = ord(data[16]) + 256*(ord(data[17])%16) val[9,0] = ord(data[18]) + 256*(ord(data[19])%16) val[10,0] = ord(data[20]) + 256*(ord(data[21])%16) val[11,0] = ord(data[22]) + 256*(ord(data[23])%16) error = ord(data[1])/16 != 0 or ord(data[3])/16 != 1 or ord(data[5])/16 != 2 or \ ord(data[7])/16 != 3 or ord(data[9])/16 != 4 or ord(data[11])/16 != 5 or\ ord(data[13])/16 != 6 or ord(data[15])/16 != 7 or ord(data[17])/16 != 8 or \ ord(data[19])/16 != 9 or ord(data[21])/16 != 10 or ord(data[23])/16 != 11 return COEFF_MAT * val, error
[ [ 1, 0, 0.2462, 0.0077, 0, 0.66, 0, 654, 0, 3, 0, 0, 654, 0, 0 ], [ 1, 0, 0.2538, 0.0077, 0, 0.66, 0.1429, 601, 0, 2, 0, 0, 601, 0, 0 ], [ 1, 0, 0.2615, 0.0077, 0, ...
[ "import time, os, copy", "import serial, numpy as np", "import threading", "from threading import RLock", "class AMTI_force:\n def __init__(self, dev_name=\"/dev/robot/force_plate0\", broadcast=True):\n self.dev_name = dev_name\n self.serial_dev = open_serial(dev_name, 57600)\n sel...
import roslib; roslib.load_manifest('hrl_lib') import rospy import std_srvs.srv as srv from hrl_msgs.msg import FloatArray import time def cb( msg ): print 'Received msg: ', msg, '\n\tType: ', msg.__class__ return x = 3.0 rospy.init_node('trial_subscriber') sub = rospy.Subscriber('trial_pub', FloatArray, cb) while not rospy.is_shutdown(): rospy.spin()
[ [ 1, 0, 0.0588, 0.0588, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0588, 0.0588, 0, 0.66, 0.1, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.1176, 0.0588, 0, 0.66,...
[ "import roslib; roslib.load_manifest('hrl_lib')", "import roslib; roslib.load_manifest('hrl_lib')", "import rospy", "import std_srvs.srv as srv", "from hrl_msgs.msg import FloatArray", "import time", "def cb( msg ):\n print('Received msg: ', msg, '\\n\\tType: ', msg.__class__)\n return", " pr...
#!/usr/bin/python import roslib roslib.load_manifest('force_torque') import rospy from geometry_msgs.msg import Vector3Stamped from threading import RLock ## 1D kalman filter update. def kalman_update(xhat, P, Q, R, z): #time update xhatminus = xhat Pminus = P + Q #measurement update K = Pminus / (Pminus + R) xhat = xhatminus + K * (z-xhatminus) P = (1-K) * Pminus return xhat, P class FTRelay: def __init__(self): self.lock = RLock() self.fresh = False def set_ft(self, value, time_acquired): self.lock.acquire() self.data = value, time_acquired self.fresh = True self.lock.release() #print 'got', value, time_acquired def get_msg(self): r = None self.lock.acquire() if self.fresh: self.fresh = False r = self.data self.lock.release() return r def FTread_to_Force( ftval, frame_id ): retval = Vector3Stamped() retval.header.stamp = rospy.rostime.get_rostime() retval.header.frame_id = frame_id retval.vector.x = ftval[0] retval.vector.y = ftval[1] retval.vector.z = ftval[2] return retval if __name__ == '__main__': import roslib; roslib.load_manifest('force_torque') import rospy from force_torque.srv import * from hrl_msgs.msg import FloatArray as FloatArray import hrl_lib.rutils as ru import time import force_torque.FTSensor as ftc import numpy as np import optparse p = optparse.OptionParser() p.add_option('--name', action='store', default='ft1', type='string', dest='name', help='name given to FTSensor') opt, args = p.parse_args() node_name = 'FTRelay_' + opt.name ft_channel_name = 'force_torque_' + opt.name service_name = node_name + '_set_ft' print node_name + ': serving service', service_name ftserver = FTRelay() rospy.init_node(node_name) rospy.Service(service_name, StringService, ru.wrap(ftserver.set_ft, ['value', 'time'], response=StringServiceResponse)) channel = rospy.Publisher(ft_channel_name, FloatArray, tcp_nodelay=True) channel2 = rospy.Publisher(ft_channel_name + '_raw', FloatArray, tcp_nodelay=True) chan_vec3 = rospy.Publisher(ft_channel_name + '_Vec3', Vector3Stamped, tcp_nodelay=True) print node_name + ': publishing on channel', ft_channel_name P_force = [1., 1., 1.] xhat_force = [0., 0., 0., 0., 0., 0.] while not rospy.is_shutdown(): msg = ftserver.get_msg() if msg is not None: data, tme = msg ftvalue = ftc.binary_to_ft(data) ftvalue = np.array(ftvalue) for i in range(3): xhat, p = kalman_update(xhat_force[i], P_force[i], 1e-3, 0.04, ftvalue[i]) P_force[i] = p xhat_force[i] = xhat #ftvalue[i] = xhat xhat_force[3] = ftvalue[3] xhat_force[4] = ftvalue[4] xhat_force[5] = ftvalue[5] ftvalue = ftvalue.tolist() channel.publish(FloatArray(rospy.Header(stamp=rospy.Time.from_seconds(tme)), xhat_force)) channel2.publish(FloatArray(rospy.Header(stamp=rospy.Time.from_seconds(tme)), ftvalue)) chan_vec3.publish( FTread_to_Force( ftvalue, opt.name )) #times.append(time.time()) #else: # time.sleep(1/5000.0) time.sleep(1/5000.0) #import pylab as pl #import numpy as np #a = np.array(times) #pl.plot(a[1:] - a[:-1], '.') #pl.show()
[ [ 1, 0, 0.0159, 0.0079, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0238, 0.0079, 0, 0.66, 0.125, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0317, 0.0079, 0, 0.6...
[ "import roslib", "roslib.load_manifest('force_torque')", "import rospy", "from geometry_msgs.msg import Vector3Stamped", "from threading import RLock", "def kalman_update(xhat, P, Q, R, z):\n #time update\n xhatminus = xhat\n Pminus = P + Q\n #measurement update\n K = Pminus / (Pminus + R)\...
#!/usr/bin/python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) # \author Cressel Anderson (Healthcare Robotics Lab, Georgia Tech.) # \author Hai Nguyen (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('force_torque') import rospy import hrl_lib.rutils as ru import hrl_lib.util as ut from hrl_msgs.msg import FloatArray from geometry_msgs.msg import WrenchStamped import numpy as np ## # Corresponding client class class FTClient(ru.GenericListener): def __init__(self, topic_name, netft=False): if netft: def msg_converter(msg): fx, fy, fz = msg.wrench.force.x, msg.wrench.force.y, msg.wrench.force.z tx, ty, tz = msg.wrench.torque.x, msg.wrench.torque.y, msg.wrench.torque.z msg_time = rospy.get_time() return -np.matrix([fx, fy, fz, tx, ty, tz]).T, msg_time msg_type=WrenchStamped else: def msg_converter(msg): m = np.matrix(msg.data, 'f').T msg_time = msg.header.stamp.to_time() return m, msg_time msg_type = FloatArray ru.GenericListener.__init__(self, 'FTClient', msg_type, topic_name, 50.0, message_extractor = msg_converter, queue_size = None) self.bias_val = np.matrix([0, 0, 0, 0, 0, 0.]).T ## # Read a force torque value # @param avg how many force torque value to average # @param without_bias # @param fresh # @param with_time_stamp # @return an averaged force torque value (6x1 matrix) def read(self, avg=1, without_bias=False, fresh=False, with_time_stamp=False): assert(avg > 0) if avg > 1: fresh = True if with_time_stamp: raise RuntimeError('Can\'t request averaging and timestamping at the same time') rs = [] for i in range(avg): if fresh: r, msg_time = ru.GenericListener.read(self, allow_duplication=False, willing_to_wait=True) else: r, msg_time = ru.GenericListener.read(self, allow_duplication=True, willing_to_wait=False) rs.append(r) readings = ut.list_mat_to_mat(rs, axis=1) if not without_bias: #print 'readiings.mean(1)', readings.mean(1) #print 'self.bias_val', self.bias_val ret = readings.mean(1) - self.bias_val else: ret = readings.mean(1) if with_time_stamp: return ret, msg_time else: return ret def bias(self): print '!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!' print 'BIASING FT' print '!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!' b_list = [] for i in range(20): r, msg_time = ru.GenericListener.read(self, allow_duplication=False, willing_to_wait=True) b_list.append(r) if b_list[0] != None: r = np.mean(np.column_stack(b_list), 1) self.bias_val = r print '!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!' print 'DONE biasing ft' print '!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!' if __name__ == '__main__': import optparse import time p = optparse.OptionParser() p.add_option('-t', action='store', default='force_torque_ft1', type='string', dest='topic', help='which topic to listen to (default force_torque_ft1)') p.add_option('--netft', action='store_true', dest='netft', help='is this a NetFT sensor') opt, args = p.parse_args() client = FTClient(opt.topic, opt.netft) client.bias() while not rospy.is_shutdown(): el = client.read() if el != None: #print np.linalg.norm(el.T) f = el.A1 print ' %.2f %.2f %.2f'%(f[0], f[1], f[2]) time.sleep(1/100.0)
[ [ 1, 0, 0.2162, 0.0068, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.2162, 0.0068, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.223, 0.0068, 0, 0.6...
[ "import roslib; roslib.load_manifest('force_torque')", "import roslib; roslib.load_manifest('force_torque')", "import rospy", "import hrl_lib.rutils as ru", "import hrl_lib.util as ut", "from hrl_msgs.msg import FloatArray", "from geometry_msgs.msg import WrenchStamped", "import numpy as np", "class...
#!/usr/bin/python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) # \author Cressel Anderson (Healthcare Robotics Lab, Georgia Tech.) # \author Hai Nguyen (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('force_torque') import rospy from hrl_msgs.msg import FloatArray #import hrl_lib.rutils as ru import AMTIForce2 as af import threading import time #import rutils as ru class AMTIForceServer(threading.Thread): def __init__(self, device_path, str_id): threading.Thread.__init__(self) try: rospy.init_node('AMTIForceServer') print 'AMTIForceServer: ros is up!' except rospy.ROSException: pass print 'AMTIForceServer: connecting to', device_path self.force_plate = af.AMTI_force(device_path) name = 'force_torque_' + str_id print 'AMTIForceServer: publishing', name, 'with type FloatArray' self.channel = rospy.Publisher(name, FloatArray, tcp_nodelay=True) def broadcast(self): print 'AMTIForceServer: started!' while not rospy.is_shutdown(): time.sleep(1/5000.0) self.channel.publish(FloatArray(None, self.force_plate.read().T.tolist()[0])) #DEPRECATED, use FTClient from ROSFTSensor with id = 0 #def AMTIForceClient(): # return ru.FloatArrayListener('AMTIForceClient', 'force_plate', 100.0) #import roslib; roslib.load_manifest('force_torque') #import force_torque.ROSAMTIForce as ft if __name__ == '__main__': import optparse p = optparse.OptionParser() p.add_option('--path', action='store', default='/dev/robot/force_plate0', type = 'string', dest='path', help='path to force torque device in (linux)') p.add_option('--name', action='store', default='ft1', type='string', dest='name', help='name given to FTSensor') opt, args = p.parse_args() server = AMTIForceServer(opt.path, opt.name) server.broadcast()
[ [ 1, 0, 0.3929, 0.0119, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.3929, 0.0119, 0, 0.66, 0.125, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.4048, 0.0119, 0, 0.6...
[ "import roslib; roslib.load_manifest('force_torque')", "import roslib; roslib.load_manifest('force_torque')", "import rospy", "from hrl_msgs.msg import FloatArray", "import AMTIForce2 as af", "import threading", "import time", "class AMTIForceServer(threading.Thread):\n def __init__(self, device_pa...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) # \author Cressel Anderson (Healthcare Robotics Lab, Georgia Tech.) # \author Hai Nguyen (Healthcare Robotics Lab, Georgia Tech.)
[]
[]
import roslib; roslib.load_manifest('hrl_lib') import rospy import std_srvs.srv as srv from hrl_msgs.msg import FloatArray import time x = 3.0 rospy.init_node('trial_publisher') pub = rospy.Publisher( 'trial_pub', FloatArray) while not rospy.is_shutdown(): x += 0.1 fa = FloatArray() fa.data = [ x, x+1.0, x+2.0 ] pub.publish( fa ) time.sleep( 0.3 )
[ [ 1, 0, 0.0625, 0.0625, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0625, 0.0625, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.125, 0.0625, 0, 0.6...
[ "import roslib; roslib.load_manifest('hrl_lib')", "import roslib; roslib.load_manifest('hrl_lib')", "import rospy", "import std_srvs.srv as srv", "from hrl_msgs.msg import FloatArray", "import time", "x = 3.0", "rospy.init_node('trial_publisher')", "pub = rospy.Publisher( 'trial_pub', FloatArray)", ...
#!/usr/bin/python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) # \author Cressel Anderson (Healthcare Robotics Lab, Georgia Tech.) # \author Hai Nguyen (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('force_torque') import rospy import hrl_lib.rutils as ru import hrl_lib.util as ut from hrl_msgs.msg import FloatArray from geometry_msgs.msg import WrenchStamped import numpy as np ## # Corresponding client class class FTClient(ru.GenericListener): def __init__(self, topic_name, netft=False): if netft: def msg_converter(msg): fx, fy, fz = msg.wrench.force.x, msg.wrench.force.y, msg.wrench.force.z tx, ty, tz = msg.wrench.torque.x, msg.wrench.torque.y, msg.wrench.torque.z msg_time = rospy.get_time() return -np.matrix([fx, fy, fz, tx, ty, tz]).T, msg_time msg_type=WrenchStamped else: def msg_converter(msg): m = np.matrix(msg.data, 'f').T msg_time = msg.header.stamp.to_time() return m, msg_time msg_type = FloatArray ru.GenericListener.__init__(self, 'FTClient', msg_type, topic_name, 50.0, message_extractor = msg_converter, queue_size = None) self.bias_val = np.matrix([0, 0, 0, 0, 0, 0.]).T ## # Read a force torque value # @param avg how many force torque value to average # @param without_bias # @param fresh # @param with_time_stamp # @return an averaged force torque value (6x1 matrix) def read(self, avg=1, without_bias=False, fresh=False, with_time_stamp=False): assert(avg > 0) if avg > 1: fresh = True if with_time_stamp: raise RuntimeError('Can\'t request averaging and timestamping at the same time') rs = [] for i in range(avg): if fresh: r, msg_time = ru.GenericListener.read(self, allow_duplication=False, willing_to_wait=True) else: r, msg_time = ru.GenericListener.read(self, allow_duplication=True, willing_to_wait=False) rs.append(r) readings = ut.list_mat_to_mat(rs, axis=1) if not without_bias: #print 'readiings.mean(1)', readings.mean(1) #print 'self.bias_val', self.bias_val ret = readings.mean(1) - self.bias_val else: ret = readings.mean(1) if with_time_stamp: return ret, msg_time else: return ret def bias(self): print '!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!' print 'BIASING FT' print '!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!' b_list = [] for i in range(20): r, msg_time = ru.GenericListener.read(self, allow_duplication=False, willing_to_wait=True) b_list.append(r) if b_list[0] != None: r = np.mean(np.column_stack(b_list), 1) self.bias_val = r print '!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!' print 'DONE biasing ft' print '!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!' if __name__ == '__main__': import optparse import time p = optparse.OptionParser() p.add_option('-t', action='store', default='force_torque_ft1', type='string', dest='topic', help='which topic to listen to (default force_torque_ft1)') p.add_option('--netft', action='store_true', dest='netft', help='is this a NetFT sensor') opt, args = p.parse_args() client = FTClient(opt.topic, opt.netft) client.bias() while not rospy.is_shutdown(): el = client.read() if el != None: #print np.linalg.norm(el.T) f = el.A1 print ' %.2f %.2f %.2f'%(f[0], f[1], f[2]) time.sleep(1/100.0)
[ [ 1, 0, 0.2162, 0.0068, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.2162, 0.0068, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.223, 0.0068, 0, 0.6...
[ "import roslib; roslib.load_manifest('force_torque')", "import roslib; roslib.load_manifest('force_torque')", "import rospy", "import hrl_lib.rutils as ru", "import hrl_lib.util as ut", "from hrl_msgs.msg import FloatArray", "from geometry_msgs.msg import WrenchStamped", "import numpy as np", "class...
#!/usr/bin/python import roslib roslib.load_manifest('force_torque') import rospy from geometry_msgs.msg import Vector3Stamped from threading import RLock ## 1D kalman filter update. def kalman_update(xhat, P, Q, R, z): #time update xhatminus = xhat Pminus = P + Q #measurement update K = Pminus / (Pminus + R) xhat = xhatminus + K * (z-xhatminus) P = (1-K) * Pminus return xhat, P class FTRelay: def __init__(self): self.lock = RLock() self.fresh = False def set_ft(self, value, time_acquired): self.lock.acquire() self.data = value, time_acquired self.fresh = True self.lock.release() #print 'got', value, time_acquired def get_msg(self): r = None self.lock.acquire() if self.fresh: self.fresh = False r = self.data self.lock.release() return r def FTread_to_Force( ftval, frame_id ): retval = Vector3Stamped() retval.header.stamp = rospy.rostime.get_rostime() retval.header.frame_id = frame_id retval.vector.x = ftval[0] retval.vector.y = ftval[1] retval.vector.z = ftval[2] return retval if __name__ == '__main__': import roslib; roslib.load_manifest('force_torque') import rospy from force_torque.srv import * from hrl_msgs.msg import FloatArray as FloatArray import hrl_lib.rutils as ru import time import force_torque.FTSensor as ftc import numpy as np import optparse p = optparse.OptionParser() p.add_option('--name', action='store', default='ft1', type='string', dest='name', help='name given to FTSensor') opt, args = p.parse_args() node_name = 'FTRelay_' + opt.name ft_channel_name = 'force_torque_' + opt.name service_name = node_name + '_set_ft' print node_name + ': serving service', service_name ftserver = FTRelay() rospy.init_node(node_name) rospy.Service(service_name, StringService, ru.wrap(ftserver.set_ft, ['value', 'time'], response=StringServiceResponse)) channel = rospy.Publisher(ft_channel_name, FloatArray, tcp_nodelay=True) channel2 = rospy.Publisher(ft_channel_name + '_raw', FloatArray, tcp_nodelay=True) chan_vec3 = rospy.Publisher(ft_channel_name + '_Vec3', Vector3Stamped, tcp_nodelay=True) print node_name + ': publishing on channel', ft_channel_name P_force = [1., 1., 1.] xhat_force = [0., 0., 0., 0., 0., 0.] while not rospy.is_shutdown(): msg = ftserver.get_msg() if msg is not None: data, tme = msg ftvalue = ftc.binary_to_ft(data) ftvalue = np.array(ftvalue) for i in range(3): xhat, p = kalman_update(xhat_force[i], P_force[i], 1e-3, 0.04, ftvalue[i]) P_force[i] = p xhat_force[i] = xhat #ftvalue[i] = xhat xhat_force[3] = ftvalue[3] xhat_force[4] = ftvalue[4] xhat_force[5] = ftvalue[5] ftvalue = ftvalue.tolist() channel.publish(FloatArray(rospy.Header(stamp=rospy.Time.from_seconds(tme)), xhat_force)) channel2.publish(FloatArray(rospy.Header(stamp=rospy.Time.from_seconds(tme)), ftvalue)) chan_vec3.publish( FTread_to_Force( ftvalue, opt.name )) #times.append(time.time()) #else: # time.sleep(1/5000.0) time.sleep(1/5000.0) #import pylab as pl #import numpy as np #a = np.array(times) #pl.plot(a[1:] - a[:-1], '.') #pl.show()
[ [ 1, 0, 0.0159, 0.0079, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0238, 0.0079, 0, 0.66, 0.125, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0317, 0.0079, 0, 0.6...
[ "import roslib", "roslib.load_manifest('force_torque')", "import rospy", "from geometry_msgs.msg import Vector3Stamped", "from threading import RLock", "def kalman_update(xhat, P, Q, R, z):\n #time update\n xhatminus = xhat\n Pminus = P + Q\n #measurement update\n K = Pminus / (Pminus + R)\...
import pygame import pygame.locals import numpy as np import time import copy #------ Slider class is code copied from the internet. (http://www.pygame.org/project/668/) class Slider(object): #Constructs the object def __init__(self, pos, value=0): self.pos = pos self.size = (275,15) self.bar = pygame.Surface((275, 15)) self.bar.fill((200, 200, 200)) self.slider = pygame.Surface((20, 15)) self.slider.fill((230, 230, 230)) pygame.draw.rect(self.bar, (0, 0, 0), (0, 0, 275, 15), 2) pygame.draw.rect(self.slider, (0, 0, 0), (-1, -1, 20, 15), 2) self.slider.set_at((19, 14), (0, 0, 0)) self.brect = self.bar.get_rect(topleft = pos) self.srect = self.slider.get_rect(topleft = pos) self.srect.left = value+pos[0] self.clicked = False self.value = value self.font_size = 15 self.font = pygame.font.SysFont("Times New Roman", self.font_size) self.text = '' def set_text(self, text): ''' set the text to be displayed below the slider ''' self.text = text #Called once every frame def update(self): mousebutton = pygame.mouse.get_pressed() cursor = pygame.locals.Rect(pygame.mouse.get_pos()[0], pygame.mouse.get_pos()[1], 1, 1) if cursor.colliderect(self.brect): if mousebutton[0]: self.clicked = True else: self.clicked = False if not mousebutton[0]: self.clicked = False if self.clicked: self.srect.center = cursor.center self.srect.clamp_ip(self.brect) self.value = self.srect.left - self.brect.left #Draws the slider def render(self, surface): surface.blit(self.bar, self.brect) surface.blit(self.slider, self.srect) ren = self.font.render(self.text,1,(0,0,0)) surface.blit(ren, (self.pos[0], self.pos[1]+self.font_size+2))
[ [ 1, 0, 0.0169, 0.0169, 0, 0.66, 0, 87, 0, 1, 0, 0, 87, 0, 0 ], [ 1, 0, 0.0339, 0.0169, 0, 0.66, 0.2, 515, 0, 1, 0, 0, 515, 0, 0 ], [ 1, 0, 0.0508, 0.0169, 0, 0.66,...
[ "import pygame", "import pygame.locals", "import numpy as np", "import time", "import copy", "class Slider(object):\n\n #Constructs the object\n def __init__(self, pos, value=0):\n self.pos = pos\n self.size = (275,15)\n\n self.bar = pygame.Surface((275, 15))", " def __init...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) # functions to use scans from hokuyos. import roslib roslib.load_manifest('hrl_hokuyo') import pygame import pygame.locals import hokuyo_scan as hs import hrl_lib.transforms as tr import hrl_lib.util as ut #import util as uto import math, numpy as np import scipy.ndimage as ni import pylab as pl import sys, time import optparse import copy import pygame_utils as pu # Need to switch to Willow's OpenCV 2.0 python bindings at some point of time. # from opencv.cv import * # from opencv.highgui import * from ctypes import POINTER MAX_DIST=1.0 # pixel coordinates of origin org_x,org_y = 320,240+150 color_list = [(255,255,0),(255,0,0),(0,255,255),(0,255,0),(0,0,255),(0,100,100),(100,100,0), (100,0,100),(100,200,100),(200,100,100),(100,100,200),(100,0,200),(0,200,100), (0,0,0),(0,100,200),(200,0,100),(100,0,100),(255,152,7) ] def angle_to_index(scan, angle): ''' returns index corresponding to angle (radians). ''' return int(min(scan.n_points-1, max(0.,int(round((angle-scan.start_angle)/scan.angular_res))))) def index_to_angle(scan, index): ''' returns angle (radians) corresponding to index. ''' return scan.start_angle+index*scan.angular_res def get_xy_map(scan,start_angle=math.radians(-360),end_angle=math.radians(360),reject_zero_ten=True,max_dist=np.Inf,min_dist=-np.Inf,sigmoid_hack=False, get_intensities=False): """ start_angle - starting angle for points to be considered (hokuyo coord) end_angle - ending angle in hokuyo coord frame. scan - object of class HokuyoScan reject_zero_ten - ignore points at 0 or 10. meters returns - 2xN matrix, if get_intensities=True returns (2xN matrix, 1xN matrix) """ start_index = angle_to_index(scan,start_angle) end_index = angle_to_index(scan,end_angle) if sigmoid_hack: #------ sigmoid hack to increase ranges of points with small range(<0.3) ---- # short_range_idxs = np.where(scan.ranges<0.3) # scan.ranges[short_range_idxs] = scan.ranges[short_range_idxs]*1.05 #--- test1 # scan.ranges = np.multiply(scan.ranges, 1+(1-np.divide(1.,1+np.power(np.e,-5*(scan.ranges-0.3))))*0.1) #--- test2 scan.ranges = np.multiply(scan.ranges, 1+(1-np.divide(1.,1+np.power(np.e,-5*(scan.ranges-0.2))))*0.20) #--- test3 # scan.ranges = np.multiply(scan.ranges, 1+(1-np.divide(1.,1+np.power(np.e,-5*(scan.ranges-0.3))))*0.2) #--- test4 # scan.ranges = np.multiply(scan.ranges, 1+(1-np.divide(1.,1+np.power(np.e,-5*(scan.ranges-0.25))))*0.15) #------------------ xy_map = np.matrix(np.row_stack((np.multiply(scan.ranges,np.cos(scan.angles)), np.multiply(scan.ranges,np.sin(scan.angles))*-1))) sub_range = scan.ranges[:,start_index:end_index+1] keep_condition = np.multiply(sub_range<=max_dist,sub_range>=min_dist) if reject_zero_ten == True: keep_condition = np.multiply(keep_condition,np.multiply(sub_range > 0.01,sub_range <= 10.)) xy_map = xy_map[:,start_index:end_index+1] idxs = np.where(keep_condition) xy_map = np.row_stack((xy_map[idxs],xy_map[idxs[0]+1,idxs[1]])) if get_intensities == True: intensities = scan.intensities[:,start_index:end_index+1] intensities = intensities[idxs] return xy_map, intensities else: return xy_map def connected_components(p, dist_threshold): ''' p - 2xN numpy matrix of euclidean points points (indices give neighbours). dist_threshold - max distance between two points in the same connected component. typical value is .02 meters returns a list of (p1,p2, p1_index, p2_index): (start euclidean point, end euclidean point, start index, end index) p1 and p2 are 2X1 matrices ''' nPoints = p.shape[1] q = p[:,0:nPoints-1] r = p[:,1:nPoints] diff = r-q dists = ut.norm(diff).T idx = np.where(dists>dist_threshold) # boundaries of the connected components end_list = idx[0].A1.tolist() end_list.append(nPoints-1) cc_list = [] start = 0 for i in end_list: cc_list.append((p[:,start], p[:,i], start, i)) start = i+1 return cc_list def find_objects(scan, max_dist, max_size, min_size, min_angle, max_angle, connect_dist_thresh, all_pts=False): ''' max_dist - objects with centroid greater than max_dist will be ignored. (meters) max_size - objects greater than this are ignored. (meters) min_size - smaller than this are ignored (meters) min_angle, max_angle - part of scan to consider. connect_dist_thresh - points in scan greater than this will be treated as separate connected components. all_pts == True: returns [ np matrix: 2xN1, 2xN2 ...] each matrix consists of all the points in the object. all_pts == False: returns list of (p,q,centroid): (end points,object centroid (2x1 matrices.)) ''' xy_map = get_xy_map(scan,min_angle,max_angle) cc_list = connected_components(xy_map,connect_dist_thresh) object_list = [] all_pts_list = [] for i,(p,q,p_idx,q_idx) in enumerate(cc_list): object_pts = xy_map[:,p_idx:q_idx+1] centroid = object_pts.sum(1)/(q_idx-p_idx+1) size = np.linalg.norm(p-q) if size>max_size: continue if size<min_size: continue if np.linalg.norm(centroid) > max_dist: continue object_list.append((p,q,centroid)) all_pts_list.append(object_pts) if all_pts == True: return all_pts_list else: return object_list def find_closest_object_point(scan, pt_interest=np.matrix([0.,0.]).T, min_angle=math.radians(-60), max_angle=math.radians(60),max_dist=0.6,min_size=0.01,max_size=0.3): ''' returns 2x1 matrix - centroid of connected component in hokuyo frame closest to pt_interest pt_interest - 2x1 matrix in hokuyo coord frame. None if no object found. ''' obj_list = find_objects(scan,max_dist,max_size,min_size,min_angle,max_angle, connect_dist_thresh=0.02, all_pts=True) if obj_list == []: return None min_dist_list = [] for pts in obj_list: min_dist_list.append(np.min(ut.norm(pts-pt_interest))) min_idx = np.argmin(np.matrix(min_dist_list)) return obj_list[min_idx].mean(1) def remove_graze_effect(ranges, angles, skip=1, graze_angle_threshold=math.radians(169.)): ''' ranges,angles - 1xN numpy matrix skip - which two rays to consider. this function changes ranges. ''' nPoints = ranges.shape[1] p = ranges[:,0:nPoints-(1+skip)] q = ranges[:,(1+skip):nPoints] d_mat = np.abs(p-q) angles_diff = np.abs(angles[:,(1+skip):nPoints]-angles[:,0:nPoints-(1+skip)]) l_mat = np.max(np.row_stack((p,q)),0) l_mat = np.multiply(l_mat,np.sin(angles_diff))/math.sin(graze_angle_threshold) thresh_exceed = d_mat>l_mat l1_greater = p>q idx_remove_1 = np.where(np.all(np.row_stack((thresh_exceed,l1_greater)),0)) idx_remove_2 = np.where(np.all(np.row_stack((thresh_exceed,1-l1_greater)),0)) # print 'idx_remove_1:', idx_remove_1 p[idx_remove_1] = 1000. q[idx_remove_2] = 1000. def remove_graze_effect_scan(scan, graze_angle_threshold=math.radians(169.)): ''' changes scan ''' remove_graze_effect(scan.ranges,scan.angles,1,graze_angle_threshold) def subtract_scans(scan2,scan1,threshold=0.01): if scan1.ranges.shape != scan2.ranges.shape: print 'hokuyo_processing.subtract_scans: the two scan.ranges have different shapes.' print 'remember to pass remove_graze_effect = False' print 'Exiting...' sys.exit() diff_range = scan2.ranges-scan1.ranges idxs = np.where(np.abs(diff_range)<threshold) #idxs = np.where(np.abs(diff_range)<0.04) # idxs = np.where(np.abs(diff_range)<0.01) # idxs = np.where(np.abs(diff_range)<0.005) hscan = hs.HokuyoScan(scan2.hokuyo_type,scan2.angular_res, scan2.max_range,scan2.min_range, scan2.start_angle,scan2.end_angle) hscan.ranges = copy.copy(scan2.ranges) hscan.ranges[idxs] = 0. return hscan def find_door(start_pts_list,end_pts_list,pt_interest=None): ''' returns [p1x,p1y], [p2x,p2y] ([],[] if no door found) returns line closest to the pt_interest. pt_interest - 2x1 matrix ''' if start_pts_list == []: return [],[] # print 'start_pts_list,end_pts_list',start_pts_list,end_pts_list start_pts = np.matrix(start_pts_list).T end_pts = np.matrix(end_pts_list).T line_vecs = end_pts-start_pts line_vecs_ang = np.arctan2(line_vecs[1,:],line_vecs[0,:]) idxs = np.where(np.add(np.multiply(line_vecs_ang>math.radians(45), line_vecs_ang<math.radians(135)), np.multiply(line_vecs_ang<math.radians(-45), line_vecs_ang>math.radians(-135)) ) > 0 )[1].A1.tolist() if idxs == []: return [],[] start_pts = start_pts[:,idxs] end_pts = end_pts[:,idxs] # print 'start_pts,end_pts',start_pts.A1.tolist(),end_pts.A1.tolist() if pt_interest == None: print 'hokuyo_processing.find_door: pt_interest in None so returning the longest line.' length = ut.norm(end_pts-start_pts) longest_line_idx = np.argmax(length) vec_door = (end_pts-start_pts)[:,longest_line_idx] return start_pts[:,longest_line_idx].A1.tolist(),end_pts[:,longest_line_idx].A1.tolist() else: v = end_pts-start_pts q_dot_v = pt_interest.T*v p1_dot_v = np.sum(np.multiply(start_pts,v),0) v_dot_v = ut.norm(v) lam = np.divide((q_dot_v-p1_dot_v),v_dot_v) r = start_pts + np.multiply(lam,v) dist = ut.norm(pt_interest-r) edge_idxs = np.where(np.multiply(lam>1.,lam<0.))[1].A1.tolist() min_end_dist = np.minimum(ut.norm(start_pts-pt_interest),ut.norm(end_pts-pt_interest)) dist[:,edge_idxs] = min_end_dist[:,edge_idxs] # dist - distance of closest point within the line segment # or distance of the closest end point. # keep line segments that are within some distance threshold. keep_idxs = np.where(dist<0.5)[1].A1.tolist() if len(keep_idxs) == 0: return [],[] start_pts = start_pts[:,keep_idxs] end_pts = end_pts[:,keep_idxs] # find distance from the robot and select furthest line. p_robot = np.matrix([0.,0.]).T v = end_pts-start_pts q_dot_v = p_robot.T*v p1_dot_v = np.sum(np.multiply(start_pts,v),0) v_dot_v = ut.norm(v) lam = np.divide((q_dot_v-p1_dot_v),v_dot_v) r = start_pts + np.multiply(lam,v) dist = ut.norm(p_robot-r) door_idx = np.argmax(dist) return start_pts[:,door_idx].A1.tolist(),end_pts[:,door_idx].A1.tolist() def xy_map_to_np_image(xy_map,m_per_pixel,dilation_count=0,padding=50): ''' returns binary numpy image. (255 for occupied pixels, 0 for unoccupied) 2d array ''' min_x = np.min(xy_map[0,:]) max_x = np.max(xy_map[0,:]) min_y = np.min(xy_map[1,:]) max_y = np.max(xy_map[1,:]) br = np.matrix([min_x,min_y]).T n_x = int(round((max_x-min_x)/m_per_pixel)) + padding n_y = int(round((max_y-min_y)/m_per_pixel)) + padding img = np.zeros((n_x+padding,n_y+padding),dtype='int') occupied_pixels = np.matrix([n_x,n_y]).T - np.round((xy_map-br)/m_per_pixel).astype('int') if dilation_count == 0: img[(occupied_pixels[0,:],occupied_pixels[1,:])] = 255 else: img[(occupied_pixels[0,:],occupied_pixels[1,:])] = 1 connect_structure = np.empty((3,3),dtype='int') connect_structure[:,:] = 1 img = ni.binary_closing(img,connect_structure,iterations=dilation_count) img = ni.binary_dilation(img,connect_structure,iterations=1) img = img*255 return img,n_x,n_y,br def xy_map_to_cv_image(xy_map,m_per_pixel,dilation_count=0,padding=10): np_im,n_x,n_y,br = xy_map_to_np_image(xy_map,m_per_pixel,dilation_count,padding) cv_im = uto.np2cv(np_im) return cv_im,n_x,n_y,br def hough_lines(xy_map,save_lines=False): ''' xy_map - 2xN matrix of points. returns start_list, end_list. [[p1x,p1y],[p2x,p2y]...],[[q1x,q1y]...] [],[] if no lines were found. ''' # save_lines=True m_per_pixel = 0.005 img_cv,n_x,n_y,br = xy_map_to_cv_image(xy_map,m_per_pixel,dilation_count=1,padding=50) time_str = str(time.time()) # for i in range(3): # cvSmooth(img_cv,img_cv,CV_GAUSSIAN,3,3) # cvSaveImage('aloha'+str(time.time())+'.png',img_cv) storage = cvCreateMemStorage(0) method = CV_HOUGH_PROBABILISTIC rho = max(int(round(0.01/m_per_pixel)),1) rho = 1 theta = math.radians(1) min_line_length = int(0.3/m_per_pixel) max_gap = int(0.1/m_per_pixel) n_points_thresh = int(0.2/m_per_pixel) # cvCanny(img_cv,img_cv,50,200) # cvSaveImage('aloha.png',img_cv) lines = cvHoughLines2(img_cv, storage, method, rho, theta, n_points_thresh, min_line_length, max_gap) if lines.total == 0: return [],[] pts_start = np.zeros((2, lines.total)) pts_end = np.zeros((2, lines.total)) if save_lines: color_dst = cvCreateImage( cvGetSize(img_cv), 8, 3 ) cvCvtColor( img_cv, color_dst, CV_GRAY2BGR ) n_lines = lines.total for idx, line in enumerate(lines.asarrayptr(POINTER(CvPoint))): pts_start[0, idx] = line[0].y pts_start[1, idx] = line[0].x pts_end[0, idx] = line[1].y pts_end[1, idx] = line[1].x if save_lines: pts_start_pixel = pts_start pts_end_pixel = pts_end pts_start = (np.matrix([n_x,n_y]).T - pts_start)*m_per_pixel + br pts_end = (np.matrix([n_x,n_y]).T - pts_end)*m_per_pixel + br along_vec = pts_end - pts_start along_vec = along_vec/ut.norm(along_vec) ang_vec = np.arctan2(-along_vec[0,:],along_vec[1,:]) res_list = [] keep_indices = [] for i in range(n_lines): ang = ang_vec[0,i] if ang>math.radians(90): ang = ang - math.radians(180) if ang<math.radians(-90): ang = ang + math.radians(180) rot_mat = tr.Rz(ang)[0:2,0:2] st = rot_mat*pts_start[:,i] en = rot_mat*pts_end[:,i] pts = rot_mat*xy_map x_all = pts[0,:] y_all = pts[1,:] min_x = min(st[0,0],en[0,0]) - 0.1 max_x = max(st[0,0],en[0,0]) + 0.1 min_y = min(st[1,0],en[1,0]) + 0.01 max_y = max(st[1,0],en[1,0]) - 0.01 keep = np.multiply(np.multiply(x_all>min_x,x_all<max_x), np.multiply(y_all>min_y,y_all<max_y)) xy_sub = xy_map[:,np.where(keep)[1].A1.tolist()] if xy_sub.shape[1] == 0: continue a,b,res = uto.fitLine_highslope(xy_sub[0,:].T, xy_sub[1,:].T) if res<0.0002: res_list.append(res) keep_indices.append(i) if keep_indices == []: return [],[] pts_start = pts_start[:,keep_indices] pts_end = pts_end[:,keep_indices] print 'number of lines:', len(keep_indices) if save_lines: ut.save_pickle(res_list,'residual_list_'+time_str+'.pkl') for i, idx in enumerate(keep_indices): s = pts_start_pixel[:,idx] e = pts_end_pixel[:,idx] cvLine(color_dst, cvPoint(int(s[1]),int(s[0])), cvPoint(int(e[1]),int(e[0])), CV_RGB(*(color_list[i])), 3, 8) cvSaveImage('lines_'+time_str+'.png',color_dst) # cvReleaseMemStorage(storage) return pts_start.T.tolist(),pts_end.T.tolist() #------------- displaying in pygame ----------- def pixel_to_real(x,y, max_dist): ''' pixel to hokuyo x,y - NX1 matrices (N points) max_dist - dist which will be drawn at row 0 ''' return (org_y-y)*max_dist/400.,(org_x-x)*max_dist/400. # return org_x-(400./max_dist)*y, org_y-(400./max_dist)*x def coord(x,y, max_dist): '''hokuyo coord frame to pixel (x,y) - floats x,y - NX1 matrices (N points) max_dist - dist which will be drawn at row 0 ''' return org_x-(400./max_dist)*y, org_y-(400./max_dist)*x def draw_points(srf,x,y,color,max_dist,step=1): ''' step - set > 1 if you don't want to draw all the points. ''' if len(x.A1) == 0: return x_pixel, y_pixel = coord(x.T,y.T,max_dist) for i in range(0,x_pixel.shape[0],step): pygame.draw.circle(srf, color, (int(x_pixel[i,0]+0.5), int(y_pixel[i,0]+0.5)), 2, 0) def draw_hokuyo_scan(srf, scan, ang1, ang2, color, reject_zero_ten=True,step=1): ''' reject_zero_ten - don't show points with 0 or 10. range readings. step - set > 1 if you don't want to draw all the points. ''' pts = get_xy_map(scan, ang1, ang2, reject_zero_ten=reject_zero_ten) # pts = get_xy_map(scan, reject_zero_ten=reject_zero_ten) max_dist = MAX_DIST draw_points(srf,pts[0,:],pts[1,:],color,max_dist,step=step) def test_connected_comps(srf, scan): #colors_list = [(200,0,0), (0,255,0), (100,100,0), (100,0,100), (0,100,100)] n_colors = len(color_list) cc_list = connected_components(get_xy_map(scan,math.radians(-60),math.radians(60)),0.03) # draw the connected components as lines for i,(p,q,p_idx,q_idx) in enumerate(cc_list): c1,c2 = p.A1.tolist(),q.A1.tolist() c1,c2 = coord(c1[0],c1[1], max_dist=MAX_DIST), coord(c2[0],c2[1], max_dist=MAX_DIST) pygame.draw.line(srf,color_list[i%n_colors],c1,c2,2) def test_find_objects(srf, scan): obj_list = find_objects(scan, max_dist=0.6, max_size=0.3, min_size=0.01, min_angle=math.radians(-60), max_angle=math.radians(60), connect_dist_thresh=0.02) print 'number of objects:', len(obj_list) #colors_list = [(200,0,0), (0,255,0), (100,100,0), (100,0,100), (0,100,100)] for i,(p,q,c) in enumerate(obj_list): if i>4: break c1,c2 = p.A1.tolist(),q.A1.tolist() c1,c2 = coord(c1[0],c1[1], max_dist=MAX_DIST), coord(c2[0],c2[1], max_dist=MAX_DIST) pygame.draw.line(srf,color_list[i],c1,c2,2) def test_find_closest_object_point(srf, scan): pt_interest = np.matrix([0.,0.]).T # pt_interest = np.matrix([0.3,-0.04]).T p = find_closest_object_point(scan, pt_interest) if p == None: return c1,c2 = coord(p[0,0],p[1,0], max_dist=MAX_DIST) pygame.draw.circle(srf, (0,200,0), (c1,c2), 3, 0) c1,c2 = coord(pt_interest[0,0],pt_interest[1,0], max_dist=MAX_DIST) pygame.draw.circle(srf, (200,200,0), (c1,c2), 3, 0) def tune_graze_effect_init(): sl = pu.Slider((340, 20), 10) return sl def tune_graze_effect_update(sl, srf, scan): sl.update() val = sl.value/255. angle = 160+val*20 remove_graze_effect_scan(scan,graze_angle_threshold=math.radians(angle)) draw_hokuyo_scan(srf,scan,math.radians(-90), math.radians(90),color=(200,0,0),reject_zero_ten=False) points_removed = np.where(np.matrix(scan.ranges)>10.)[0].shape[1] sl.set_text('angle: %.2f, points_removed: %d'%(angle, points_removed)) sl.render(srf) if sl.clicked: return True else: return False def test_find_handle_init(): fr = np.matrix([1.28184669,0.05562259]).T bl = np.matrix([1.19585711,-0.06184923]).T max_dist = MAX_DIST x_fr, y_fr = coord(fr[0,0],fr[1,0],max_dist) x_bl, y_bl = coord(bl[0,0],bl[1,0],max_dist) pygame.draw.rect(srf,(200,0,200),pygame.Rect(x_bl,y_bl,x_fr-x_bl,y_fr-y_bl),1) cv_handle_template = create_handle_template(dict['scan'],dict['bk_lt'],dict['fr_rt']) def test_find_lines(): pts = get_xy_map(scan,math.radians(-60),math.radians(60)) p_start_list,p_end_list = hough_lines(pts) if p_start_list == []: return #------- to test door finding ---------- # p_start,p_end = find_door(p_start_list,p_end_list) # if p_start == []: # return # p_start_list,p_end_list = [p_start],[p_end] n_colors = len(color_list) for i,(p1,p2) in enumerate(zip(p_start_list,p_end_list)): c1,c2 = coord(p1[0],p1[1], max_dist=MAX_DIST), coord(p2[0],p2[1], max_dist=MAX_DIST) pygame.draw.line(srf,color_list[i%n_colors],c1,c2,2) if __name__ == '__main__': p = optparse.OptionParser() p.add_option('-t', action='store', type='string', dest='hokuyo_type', help='hokuyo_type. urg or utm') p.add_option('-a', action='store', type='int', dest='avg', help='number of scans to average', default=1) p.add_option('-n', action='store', type='int', dest='hokuyo_number', help='hokuyo number. 0,1,2 ...') p.add_option('-f', action='store_true', dest='flip', help='flip the hokuyo scan') p.add_option('--ang_range', type='float', dest='ang_range', help='max angle of the ray to display (degrees)', default=360.) opt, args = p.parse_args() hokuyo_type = opt.hokuyo_type hokuyo_number = opt.hokuyo_number avg_number = opt.avg flip = opt.flip ang_range = opt.ang_range ang_range = math.radians(ang_range) #------- which things to test --------- test_graze_effect_flag = False test_find_objects_flag = False test_find_closest_object_point_flag = False test_show_change_flag = False test_find_lines_flag = False #-------------------------------------- # Initialize pygame # pygame.init() # Open a display srf = pygame.display.set_mode((640,480)) pygame.display.set_caption(hokuyo_type+' '+str(hokuyo_number)) fps = 100 loopFlag = True clk = pygame.time.Clock() if hokuyo_type == 'utm': h = hs.Hokuyo('utm',hokuyo_number,start_angle=-ang_range, end_angle=ang_range,flip=flip) elif hokuyo_type == 'urg': h = hs.Hokuyo('urg',hokuyo_number,flip=flip) else: print 'unknown hokuyo type: ', hokuyo_type print 'Exiting...' sys.exit() # scan1 = h.get_scan(avoid_duplicate=True) # sys.exit() #---------- initializations ------------- if test_graze_effect_flag: sl = tune_graze_effect_init() #--------- and now loop ----------- if test_show_change_flag: scan_prev = h.get_scan(avoid_duplicate=True, avg=avg_number, remove_graze=False) n_ch_pts_list = [] while loopFlag: widget_clicked = False # Clear the screen srf.fill((255,255,255)) # Draw the urg pygame.draw.circle(srf, (200,0,0), (org_x,org_y), 10, 0) #display all the points if test_graze_effect_flag or test_show_change_flag: # here we don't want any filtering on the scan. scan = h.get_scan(avoid_duplicate=True, avg=avg_number, remove_graze=False) else: scan = h.get_scan(avoid_duplicate=True, avg=avg_number, remove_graze=True) if test_show_change_flag == False: #draw_hokuyo_scan(srf,scan,math.radians(-70), math.radians(70),color=(0,0,200)) #draw_hokuyo_scan(srf,scan,math.radians(-90), math.radians(90),color=(0,0,200)) draw_hokuyo_scan(srf,scan,math.radians(-135),math.radians(135),color=(0,0,200)) else: diff_scan = subtract_scans(scan,scan_prev) scan_prev = scan pts = get_xy_map(diff_scan,math.radians(-40), math.radians(40),reject_zero_ten=True) n_ch_pts_list.append(pts.shape[1]) max_dist = MAX_DIST draw_points(srf,pts[0,:],pts[1,:],color=(0,200,0),max_dist=max_dist) # test_connected_comps(srf, scan) if test_find_objects_flag: test_find_objects(srf, scan) if test_find_closest_object_point_flag: test_find_closest_object_point(srf, scan) if test_graze_effect_flag: widget_clicked |= tune_graze_effect_update(sl,srf, scan) if test_find_lines_flag: test_find_lines() pygame.display.flip() events = pygame.event.get() for e in events: if e.type==pygame.locals.QUIT: loopFlag=False if e.type==pygame.locals.KEYDOWN: if e.key == 27: # Esc loopFlag=False if widget_clicked == False: if e.type==pygame.locals.MOUSEMOTION: if e.buttons[0] == 1: # left button org_x += e.rel[0] org_y += e.rel[1] if e.buttons[2] == 1: # right button MAX_DIST *= (1+e.rel[1]*0.01) MAX_DIST = max(MAX_DIST,0.1) # Try to keep the specified framerate clk.tick(fps) if test_show_change_flag: ut.save_pickle(n_ch_pts_list,ut.formatted_time()+'_ch_pts_list.pkl')
[ [ 1, 0, 0.0477, 0.0014, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0491, 0.0014, 0, 0.66, 0.0238, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0519, 0.0014, 0, 0....
[ "import roslib", "roslib.load_manifest('hrl_hokuyo')", "import pygame", "import pygame.locals", "import hokuyo_scan as hs", "import hrl_lib.transforms as tr", "import hrl_lib.util as ut", "import math, numpy as np", "import scipy.ndimage as ni", "import pylab as pl", "import sys, time", "impor...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import math, numpy as np import sys, time NAME = 'utm_python_listener' import roslib; roslib.load_manifest('hrl_hokuyo') import rospy from sensor_msgs.msg import LaserScan from threading import RLock import hokuyo_processing as hp import copy import numpy as np class HokuyoScan(): ''' This class has the data of a laser scan. ''' def __init__(self,hokuyo_type,angular_res,max_range,min_range,start_angle,end_angle): ''' hokuyo_type - 'urg', 'utm' angular_res - angle (radians) between consecutive points in the scan. min_range, max_range - in meters ranges,angles,intensities - 1xn_points numpy matrix. ''' self.hokuyo_type = hokuyo_type self.angular_res = angular_res self.max_range = max_range self.min_range = min_range self.start_angle = start_angle self.end_angle = end_angle self.n_points = int((end_angle-start_angle)/angular_res)+1 self.ranges = None self.intensities = None self.angles = [] for i in xrange(self.n_points): self.angles.append(hp.index_to_angle(self,i)) self.angles = np.matrix(self.angles) class Urg(): ''' get a scan from urg using player. ''' def __init__(self, urg_number,start_angle=None,end_angle=None, host='localhost', port=6665): ''' host, port - hostname and port of the player server ''' import player_functions as pf import playerc as pl self.player_client = pf.player_connect(host,port) self.laser = pl.playerc_laser(self.player_client, urg_number) if self.laser.subscribe(pl.PLAYERC_OPEN_MODE) != 0: raise RuntimeError("hokuyo_scan.Urg.__init__: Unable to connect to urg!\n") for i in range(10): self.player_client.read() start_angle_fullscan = -math.radians(654./2*0.36) end_angle_fullscan = math.radians(654./2*0.36) if start_angle == None or start_angle<start_angle_fullscan: start_angle_subscan = start_angle_fullscan else: start_angle_subscan = start_angle if end_angle == None or end_angle>end_angle_fullscan: end_angle_subscan = end_angle_fullscan else: end_angle_subscan = end_angle ang_res = math.radians(0.36) self.start_index_fullscan=(start_angle_subscan-start_angle_fullscan)/ang_res self.end_index_fullscan=(end_angle_subscan-start_angle_fullscan)/ang_res self.hokuyo_scan = HokuyoScan(hokuyo_type = 'urg', angular_res = math.radians(0.36), max_range = 4.0, min_range = 0.2, start_angle=start_angle_subscan, end_angle=end_angle_subscan) def get_scan(self, avoid_duplicate=False): ''' avoid_duplicate - avoids duplicates caused by querying player faster than it can get new scans. ''' curr_ranges = self.hokuyo_scan.ranges for i in range(10): # I get a new scan by the time i=4 if self.player_client.read() == None: raise RuntimeError('hokuyo_scan.Urg.get_scan: player_client.read() returned None.\n') sub_ranges = np.matrix(self.laser.ranges[self.start_index_fullscan:self.end_index_fullscan+1]) if avoid_duplicate == False or np.any(curr_ranges != sub_ranges): # got a fresh scan from player. break self.hokuyo_scan.ranges = sub_ranges return copy.copy(self.hokuyo_scan) class Utm(): ''' get a scan from a UTM using ROS ''' def __init__(self, utm_number,start_angle=None,end_angle=None,ros_init_node=True): hokuyo_node_name = '/utm%d'%utm_number # max_ang = rospy.get_param(hokuyo_node_name+'/max_ang') # min_ang = rospy.get_param(hokuyo_node_name+'/min_ang') # start_angle_fullscan = min_ang # end_angle_fullscan = max_ang # print 'max_angle:', math.degrees(max_ang) # print 'min_angle:', math.degrees(min_ang) max_ang_degrees = rospy.get_param(hokuyo_node_name+'/max_ang_degrees') min_ang_degrees = rospy.get_param(hokuyo_node_name+'/min_ang_degrees') start_angle_fullscan = math.radians(min_ang_degrees) end_angle_fullscan = math.radians(max_ang_degrees) # This is actually determined by the ROS node params and not the UTM. # start_angle_fullscan = -math.radians(1080./2*0.25) #270deg # end_angle_fullscan = math.radians(1080./2*0.25) # start_angle_fullscan = -math.radians(720./2*0.25) #180deg # end_angle_fullscan = math.radians(720./2*0.25) # start_angle_fullscan = -math.radians(559./2*0.25) #140deg # end_angle_fullscan = math.radians(559./2*0.25) if start_angle == None or start_angle<start_angle_fullscan: start_angle_subscan = start_angle_fullscan else: start_angle_subscan = start_angle if end_angle == None or end_angle>end_angle_fullscan: end_angle_subscan = end_angle_fullscan else: end_angle_subscan = end_angle ang_res = math.radians(0.25) self.start_index_fullscan=int(round((start_angle_subscan-start_angle_fullscan)/ang_res)) self.end_index_fullscan=int(round((end_angle_subscan-start_angle_fullscan)/ang_res)) self.hokuyo_scan = HokuyoScan(hokuyo_type = 'utm', angular_res = math.radians(0.25), max_range = 10.0, min_range = 0.1, start_angle=start_angle_subscan, end_angle=end_angle_subscan) self.lock = RLock() self.lock_init = RLock() self.connected_to_ros = False self.ranges,self.intensities = None,None if ros_init_node: try: print 'Utm: init ros node.' rospy.init_node(NAME, anonymous=True) except rospy.ROSException, e: print 'Utm: rospy already initialized. Got message', e pass rospy.Subscriber("utm%d_scan"%(utm_number), LaserScan, self.callback, queue_size = 1) def callback(self, scan): self.lock.acquire() self.connected_to_ros = True self.ranges = np.matrix(scan.ranges[self.start_index_fullscan:self.end_index_fullscan+1]) self.intensities = np.matrix(scan.intensities[self.start_index_fullscan:self.end_index_fullscan+1]) self.lock.release() def get_scan(self, avoid_duplicate=False): while self.connected_to_ros == False: pass while not rospy.is_shutdown(): self.lock.acquire() if avoid_duplicate == False or np.any(self.hokuyo_scan.ranges!=self.ranges): # got a fresh scan from ROS self.hokuyo_scan.ranges = copy.copy(self.ranges) self.hokuyo_scan.intensities = copy.copy(self.intensities) self.lock.release() break self.lock.release() time.sleep(0.001) return copy.copy(self.hokuyo_scan) class Hokuyo(): ''' common class for getting scans from both urg and utm. ''' def __init__(self, hokuyo_type, hokuyo_number, start_angle=None, end_angle=None, flip=False, ros_init_node=True): ''' hokuyo_type - 'urg', 'utm' - 'utm' requires ros revision 2536 ros-kg revision 5612 hokuyo_number - 0,1,2 ... start_angle, end_angle - unoccluded part of the scan. (radians) flip - whether to flip the scans (hokuyo upside-down) ''' self.hokuyo_type = hokuyo_type self.flip = flip if hokuyo_type == 'urg': self.hokuyo = Urg(hokuyo_number,start_angle,end_angle) elif hokuyo_type == 'utm': self.hokuyo = Utm(hokuyo_number,start_angle,end_angle,ros_init_node=ros_init_node) else: raise RuntimeError('hokuyo_scan.Hokuyo.__init__: Unknown hokuyo type: %s\n'%(hokuyo_type)) def get_scan(self, avoid_duplicate=False, avg=1, remove_graze=True): ''' avoid_duplicate - prevent duplicate scans which will happen if get_scan is called at a rate faster than the scanning rate of the hokuyo. avoid_duplicate == True, get_scan will block till new scan is received. (~.2s for urg and 0.05s for utm) ''' l = [] l2 = [] for i in range(avg): hscan = self.hokuyo.get_scan(avoid_duplicate) l.append(hscan.ranges) l2.append(hscan.intensities) ranges_mat = np.row_stack(l) ranges_mat[np.where(ranges_mat==0)] = -1000. # make zero pointvery negative ranges_avg = (ranges_mat.sum(0)/avg) if self.flip: ranges_avg = np.fliplr(ranges_avg) hscan.ranges = ranges_avg intensities_mat = np.row_stack(l2) if self.hokuyo_type == 'utm': hscan.intensities = (intensities_mat.sum(0)/avg) if remove_graze: if self.hokuyo_type=='utm': hp.remove_graze_effect_scan(hscan) else: print 'hokuyo_scan.Hokuyo.get_scan: hokuyo type is urg, but remove_graze is True' return hscan if __name__ == '__main__': import pylab as pl # h = Hokuyo('urg', 0) h = Hokuyo('utm', 0, flip=True) print 'getting first scan' scan1 = h.get_scan(avoid_duplicate=True) # hp.get_xy_map(scan1) t_list = [] for i in range(200): t0 = time.time() scan2 = h.get_scan(avoid_duplicate=True,avg=1,remove_graze=False) # scan = h.get_scan(avoid_duplicate=False) t1 = time.time() print t1-t0, scan1==scan2 t_list.append(t1-t0) scan1=scan2 print scan1.ranges.shape print scan1.angles.shape # t_mat = np.matrix(t_list) # print '#################################################' # print 'mean time between scans:', t_mat.mean() # print 'standard deviation:', np.std(t_mat) pl.plot(t_list) pl.show()
[ [ 1, 0, 0.0055, 0.0055, 0, 0.66, 0, 526, 0, 2, 0, 0, 526, 0, 0 ], [ 1, 0, 0.011, 0.0055, 0, 0.66, 0.0769, 509, 0, 2, 0, 0, 509, 0, 0 ], [ 1, 0, 0.022, 0.0055, 0, 0....
[ "import math, numpy as np", "import sys, time", "import roslib; roslib.load_manifest('hrl_hokuyo')", "import roslib; roslib.load_manifest('hrl_hokuyo')", "import rospy", "from sensor_msgs.msg import LaserScan", "from threading import RLock", "import hokuyo_processing as hp", "import copy", "import...
import math # home position in encoder ticks for the servo. servo_param = { # 1: { # Default for new servo. Please issue 'new_servo.write_id(new_id)' and setup your own home position! # 'home_encoder': 351 # }, 2: { # Tilting Hokuyo on El-E 'home_encoder': 446 }, 3: { # RFID Antenna Left Tilt 'home_encoder': 377 }, 4: { # RFID Antenna Right Tilt 'home_encoder': 330 }, 5: { # Tilting kinect on Cody 'home_encoder': 447, 'max_ang': math.radians(55.), 'min_ang': math.radians(-80.) }, 6: { # EL-E stereohead Pan 'home_encoder': 500, 'max_ang': math.radians(90.), 'min_ang': math.radians(-90.) }, 7: { # EL-E safetyscreen tilt. 'home_encoder': 373 }, 11: { # RFID Left Pan 'home_encoder': 430, 'max_ang': math.radians( 141.0 ), 'min_ang': math.radians( -31.0 ) }, 12: { # RFID Left Tilt 'home_encoder': 507, 'max_ang': math.radians( 46.0 ), 'min_ang': math.radians( -36.0 ) }, 13: { # RFID Right Pan 'home_encoder': 583, 'max_ang': math.radians( 31.0 ), 'min_ang': math.radians( -141.0 ) }, 14: { # RFID Right Tilt 'home_encoder': 504, 'max_ang': math.radians( 46.0 ), 'min_ang': math.radians( -36.0 ) }, 15: { # Ear Flap on RFID El-E Right 'home_encoder': 498 }, 16: { # Pan Antenna on RFID El-E Left 'home_encoder': 365 }, 17: { # Tilt Antenna on RFID El-E Left 'home_encoder': 504 }, 18: { # EL-E stereohead Tilt 'home_encoder': 495, 'max_ang': math.radians(60.), 'min_ang': math.radians(-20.) }, 19: { # Desktop System UTM 'home_encoder': 633, 'flipped': True }, 20: { # Desktop System Tilt 'home_encoder': 381 }, 21: { # Desktop System Pan 'home_encoder': 589 }, 22: { # Desktop System Roll 'home_encoder': 454 }, 23: { # Dinah Top 'home_encoder': 379 }, 24: { # Dinah Bottom 'home_encoder': 365 }, 25: { # Meka top Pan 'home_encoder': 500 }, 26: { # Meka top Tilt 'home_encoder': 400 }, 27: { # PR2 RFID Right Pan 'home_encoder': 512 }, 28: { # PR2 RFID Right Tilt 'home_encoder': 506 }, 29: { # PR2 RFID Left Pan 'home_encoder': 544 }, 30: { # PR2 RFID Left Tilt 'home_encoder': 500 }, 31: { # Playpen open/close 'home_encoder': 381 }, 32: { # Conveyor for playpen 'home_encoder': 1 } }
[ [ 1, 0, 0.0092, 0.0092, 0, 0.66, 0, 526, 0, 1, 0, 0, 526, 0, 0 ], [ 14, 0, 0.5183, 0.9541, 0, 0.66, 1, 241, 0, 0, 0, 0, 0, 6, 14 ] ]
[ "import math", "servo_param = {\n# 1: { # Default for new servo. Please issue 'new_servo.write_id(new_id)' and setup your own home position!\n# 'home_encoder': 351\n# }, \n 2: { # Tilting Hokuyo on El-E\n 'home_encoder': 446\n }, \n ...
__all__ = [ 'robotis_servo', 'lib_robotis', 'ros_robotis', 'servo_config' ]
[ [ 14, 0, 0.5833, 1, 0, 0.66, 0, 272, 0, 0, 0, 0, 0, 5, 0 ] ]
[ "__all__ = [\n'robotis_servo',\n'lib_robotis',\n'ros_robotis',\n'servo_config'\n]" ]
#!/usr/bin/python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # ## Controlling Robotis Dynamixel RX-28 & RX-64 servos from python ## using the USB2Dynamixel adaptor. ## Authors: Travis Deyle & Advait Jain (Healthcare Robotics Lab, Georgia Tech.) # ROS imports import roslib roslib.load_manifest('robotis') import rospy from std_msgs.msg import Float64 from robotis.srv import None_Float from robotis.srv import None_FloatResponse from robotis.srv import MoveAng from robotis.srv import MoveAngResponse from robotis.srv import None_Int32 from robotis.srv import None_Int32Response import robotis.lib_robotis as rs import time import math from threading import Thread class ROS_Robotis_Server(): # This class provides ROS services for a select few lib_robotis functions def __init__(self, servo = None, name = '' ): if servo == None: raise RuntimeError( 'ROS_Robotis_Servo: No servo specified for server.\n' ) self.servo = servo self.name = name try: rospy.init_node( 'robotis_servo_' + self.name ) except rospy.ROSException: pass #self.servo.disable_torque() rospy.logout( 'ROS_Robotis_Servo: Starting Server /robotis/servo_' + self.name ) self.channel = rospy.Publisher('/robotis/servo_' + self.name, Float64) self.__service_ang = rospy.Service('/robotis/servo_' + name + '_readangle', None_Float, self.__cb_readangle) self.__service_ismove = rospy.Service('/robotis/servo_' + name + '_ismoving', None_Int32, self.__cb_ismoving) self.__service_moveang = rospy.Service('/robotis/servo_' + name + '_moveangle', MoveAng, self.__cb_moveangle) def __cb_readangle( self, request ): ang = self.update_server() return None_FloatResponse( ang ) def __cb_ismoving( self, request ): status = self.servo.is_moving() return None_Int32Response( int(status) ) def __cb_moveangle( self, request ): ang = request.angle angvel = request.angvel blocking = bool( request.blocking ) self.servo.move_angle( ang, angvel, blocking ) return MoveAngResponse() def update_server(self): ang = self.servo.read_angle() self.channel.publish( Float64(ang) ) return ang class ROS_Robotis_Poller( Thread ): # A utility class that will setup and poll a number of ROS_Robotis_Servos on one USB2Dynamixel def __init__( self, dev_name, ids, names ): Thread.__init__(self) self.should_run = True self.dev_name = dev_name self.ids = ids self.names = names for n in self.names: rospy.logout( 'ROS_Robotis_Servo: Starting Up /robotis/servo_' + n + ' on ' + self.dev_name ) self.dyn = rs.USB2Dynamixel_Device( self.dev_name ) self.servos = [ rs.Robotis_Servo( self.dyn, i ) for i in self.ids ] self.ros_servers = [ ROS_Robotis_Server( s, n ) for s,n in zip( self.servos, self.names ) ] rospy.logout( 'ROS_Robotis_Servo: Setup Complete on ' + self.dev_name ) self.start() def run( self ): while self.should_run and not rospy.is_shutdown(): [ s.update_server() for s in self.ros_servers ] time.sleep(0.001) for n in self.names: rospy.logout( 'ROS_Robotis_Servo: Shutting Down /robotis/servo_' + n + ' on ' + self.dev_name ) def stop(self): self.should_run = False self.join(3) if (self.isAlive()): raise RuntimeError("ROS_Robotis_Servo: unable to stop thread") class ROS_Robotis_Client(): # Provides access to the ROS services in the server. def __init__(self, name = '' ): self.name = name rospy.wait_for_service('/robotis/servo_' + name + '_readangle') rospy.wait_for_service('/robotis/servo_' + name + '_ismoving') rospy.wait_for_service('/robotis/servo_' + name + '_moveangle') self.__service_ang = rospy.ServiceProxy('/robotis/servo_' + name + '_readangle', None_Float) self.__service_ismove = rospy.ServiceProxy('/robotis/servo_' + name + '_ismoving', None_Int32) self.__service_moveang = rospy.ServiceProxy('/robotis/servo_' + name + '_moveangle', MoveAng) def read_angle( self ): resp = self.__service_ang() ang = resp.value return ang def is_moving( self ): resp = self.__service_ismove() return bool( resp.value ) def move_angle( self, ang, angvel = math.radians(50), blocking = True ): self.__service_moveang( ang, angvel, int(blocking) ) if __name__ == '__main__': print 'Sample Server: ' # Important note: You cannot (!) use the same device (dyn) in another # process. The device is only "thread-safe" within the same # process (i.e. between servos (and callbacks) instantiated # within that process) # NOTE: If you are going to be polling the servers as in the snippet # below, I recommen using a poller! See "SAMPLE POLLER" below. dev_name = '/dev/robot/servo_left' ids = [11, 12] names = ['pan', 'tilt'] dyn = rs.USB2Dynamixel_Device( dev_name ) servos = [ rs.Robotis_Servo( dyn, i ) for i in ids ] ros_servers = [ ROS_Robotis_Server( s, n ) for s,n in zip( servos, names ) ] try: while not rospy.is_shutdown(): [ s.update_server() for s in ros_servers ] time.sleep(0.001) except: pass for n in names: print 'ROS_Robotis_Servo: Shutting Down /robotis/servo_'+n ## SAMPLE POLLER # The above example just constantly polls all the servos, while also # making the services available. This generally useful code is # encapsulated in a more general poller class (which also has nicer # shutdown / restart properties). Thus, the above example is best used as: # ROS_Robotis_Poller( '/dev/robot/servo_left', [11,12], ['pan', 'tilt'] ) ## SAMPLE CLIENTS: # tilt = ROS_Robotis_Client( 'tilt' ) # tilt.move_angle( math.radians( 0 ), math.radians(10), blocking=False) # while tilt.is_moving(): # print 'Tilt is moving' # pan = ROS_Robotis_Client( 'pan' ) # pan.move_angle( math.radians( 0 ), math.radians(10), blocking=False) # while pan.is_moving(): # print 'pan is moving'
[ [ 1, 0, 0.1629, 0.0045, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.1674, 0.0045, 0, 0.66, 0.0588, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.1719, 0.0045, 0, 0....
[ "import roslib", "roslib.load_manifest('robotis')", "import rospy", "from std_msgs.msg import Float64", "from robotis.srv import None_Float", "from robotis.srv import None_FloatResponse", "from robotis.srv import MoveAng", "from robotis.srv import MoveAngResponse", "from robotis.srv import None_Int3...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('pan_tilt_robotis') import time import sys, optparse import numpy as np, math import hrl_lib.util as ut import robotis.robotis_servo as rs class PanTilt(): ## Assumes that both pan and tilt servos are controlled using the # same usb2dynamixel adaptor. # @param dev_name - name of serial device of the servo controller (e.g. '/dev/robot/servo0') # @param pan_id - servo id for the pan Robotis servo. # @param pan_id - servo id for the tilt Robotis servo. # @param baudrate - for the servo controller (usb2dynamixel) # @param pan_speed - max pan speed (radians/sec) # @param tilt_speed - max tilt speed (radians/sec) def __init__(self, dev_name, pan_id, tilt_id, baudrate=57600, pan_speed = math.radians(180), tilt_speed = math.radians(180)): self.pan_servo = rs.robotis_servo(dev_name,pan_id,baudrate, max_speed = pan_speed) self.tilt_servo = rs.robotis_servo(dev_name,tilt_id,baudrate, max_speed = tilt_speed) self.max_pan = self.pan_servo.max_ang self.min_pan = self.pan_servo.min_ang self.max_tilt = self.tilt_servo.max_ang self.min_tilt = self.tilt_servo.min_ang ## return (pan,tilt) angles in RADIANS. def get_pan_tilt(self): pan = self.pan_servo.read_angle() tilt = self.tilt_servo.read_angle() return pan, -tilt ## set (pan,tilt) angles in RADIANS. # blocks until the pan and tilt angles are attained. # @param pan - pan angle (RADIANS) # @param tilt - tilt angle (RADIANS) def set_pan_tilt(self, pan, tilt, speed=math.radians(180)): self.pan_servo.move_angle(pan, angvel=speed, blocking=False) self.tilt_servo.move_angle(tilt, angvel=speed, blocking=True) self.pan_servo.move_angle(pan, angvel=speed, blocking=True) ## new pan,tilt = current pan,tilt + pan_d,tilt_d # blocks until the pan and tilt angles are attained. # @param pan - pan angle (RADIANS) # @param tilt - tilt angle (RADIANS) def set_pan_tilt_delta(self,pan_d,tilt_d): p,t = self.get_pan_tilt() self.set_pan_tilt(p+pan_d,t+tilt_d) def set_ptz_angles_rad(self, pan, tilt): print 'pan_tilt.set_ptz_angles_rad: WARNING this function has been deprecated. use set_pan_tilt' self.set_pan_tilt(pan, tilt) def set_ptz_values(self, pan, tilt, blocking=True): print 'pan_tilt.set_ptz_values: WARNING this function has been deprecated. use set_pan_tilt' self.set_pan_tilt(pan, tilt) def get_ptz_angles_rad(self): print 'pan_tilt.get_ptz_angles_rad: WARNING this function has been deprecated. use set_pan_tilt' return self.get_pan_tilt() def get_ptz_values(self): print 'pan_tilt.get_ptz_values: WARNING this function has been deprecated. use set_pan_tilt' p, t = self.get_pan_tilt() #return p, t return math.degrees(p), math.degrees(t) if __name__ == '__main__': p = optparse.OptionParser() p.add_option('-d', action='store', type='string', dest='servo_dev_name', default='/dev/robot/pan_tilt0', help='servo device string. [default= /dev/robot/pan_tilt0]') p.add_option('--pan_id', action='store', type='int', dest='pan_id', help='id of the pan servo',default=None) p.add_option('--tilt_id', action='store', type='int', dest='tilt_id', help='id of the tilt servo',default=None) p.add_option('--pan', action='store', type='float', dest='pan', help='pan angle (degrees).',default=None) p.add_option('--tilt', action='store', type='float', dest='tilt', help='tilt angle (degrees).',default=None) opt, args = p.parse_args() servo_dev_name = opt.servo_dev_name pan_id = opt.pan_id tilt_id = opt.tilt_id pan = opt.pan tilt = opt.tilt if pan_id == None: print 'Please provide a pan_id' print 'Exiting...' sys.exit() if tilt_id == None: print 'Please provide a tilt_id' print 'Exiting...' sys.exit() if pan == None: print 'Please provide a pan (angle)' print 'Exiting...' sys.exit() if tilt == None: print 'Please provide a tilt (angle)' print 'Exiting...' sys.exit() ptu = PanTilt(servo_dev_name,pan_id,tilt_id) ptu.set_pan_tilt(math.radians(pan),math.radians(tilt)) # For EL-E: # python pan_tilt.py -d /dev/robot/servos_pan_tilt_hat --pan_id=6 --tilt_id=18 --pan=0 --tilt=0
[ [ 1, 0, 0.1923, 0.0064, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.1923, 0.0064, 0, 0.66, 0.125, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.2051, 0.0064, 0, 0.6...
[ "import roslib; roslib.load_manifest('pan_tilt_robotis')", "import roslib; roslib.load_manifest('pan_tilt_robotis')", "import time", "import sys, optparse", "import numpy as np, math", "import hrl_lib.util as ut", "import robotis.robotis_servo as rs", "class PanTilt():\n \n ## Assumes that both ...
import time import sys, os, copy import numpy as np, math import scipy.ndimage as ni class occupancy_grid_3d(): ## # @param resolution - 3x1 matrix. size of each cell (in meters) along # the different directions. def __init__(self, center, size, resolution, data, occupancy_threshold, to_binary = True): self.grid_shape = size/resolution tlb = center + size/2 brf = center + size/2 self.size = size self.grid = np.reshape(data, self.grid_shape) self.grid_shape = np.matrix(self.grid.shape).T self.resolution = resolution self.center = center if to_binary: self.to_binary(occupancy_threshold) ## binarize the grid # @param occupancy_threshold - voxels with occupancy less than this are set to zero. def to_binary(self, occupancy_threshold): filled = (self.grid >= occupancy_threshold) self.grid[np.where(filled==True)] = 1 self.grid[np.where(filled==False)] = 0 ## # @param array - if not None then this will be used instead of self.grid # @return 3xN matrix of 3d coord of the cells which have occupancy = 1 def grid_to_points(self, array=None): if array == None: array = self.grid idxs = np.where(array == 1) x_idx = idxs[0] y_idx = idxs[1] z_idx = idxs[2] x = x_idx * self.resolution[0,0] + self.center[0,0] - self.size[0,0]/2 y = y_idx * self.resolution[1,0] + self.center[1,0] - self.size[1,0]/2 z = z_idx * self.resolution[2,0] + self.center[2,0] - self.size[2,0]/2 return np.matrix(np.row_stack([x,y,z])) ## 27-connected components. # @param threshold - min allowed size of connected component def connected_comonents(self, threshold): connect_structure = np.ones((3,3,3), dtype='int') grid = self.grid labeled_arr, n_labels = ni.label(grid, connect_structure) if n_labels == 0: return labeled_arr, n_labels labels_list = range(1,n_labels+1) count_objects = ni.sum(grid, labeled_arr, labels_list) if n_labels == 1: count_objects = [count_objects] t0 = time.time() new_labels_list = [] for c,l in zip(count_objects, labels_list): if c > threshold: new_labels_list.append(l) else: labeled_arr[np.where(labeled_arr == l)] = 0 # relabel stuff for nl,l in enumerate(new_labels_list): labeled_arr[np.where(labeled_arr == l)] = nl+1 n_labels = len(new_labels_list) t1 = time.time() print 'time:', t1-t0 return labeled_arr, n_labels if __name__ == '__main__': print 'Hello World'
[ [ 1, 0, 0.0225, 0.0112, 0, 0.66, 0, 654, 0, 1, 0, 0, 654, 0, 0 ], [ 1, 0, 0.0337, 0.0112, 0, 0.66, 0.2, 509, 0, 3, 0, 0, 509, 0, 0 ], [ 1, 0, 0.0449, 0.0112, 0, 0.6...
[ "import time", "import sys, os, copy", "import numpy as np, math", "import scipy.ndimage as ni", "class occupancy_grid_3d():\n\n ##\n # @param resolution - 3x1 matrix. size of each cell (in meters) along\n # the different directions.\n def __init__(self, center, size, resolut...
import sys import numpy as np, math import add_cylinder as ac import online_collision_detection as ocd import roslib; roslib.load_manifest('arm_navigation_tutorials') import rospy from mapping_msgs.msg import CollisionObject from visualization_msgs.msg import Marker import hrl_lib.transforms as tr import hrl_lib.viz as hv roslib.load_manifest('hrl_pr2_lib') import hrl_pr2_lib.pr2_arms as pa roslib.load_manifest('force_torque') # hack by Advait import force_torque.FTClient as ftc import tf class object_ft_sensors(): def __init__(self): self.obj1_ftc = ftc.FTClient('force_torque_ft2') self.tf_lstnr = tf.TransformListener() def get_forces(self, bias = True): # later I might be looping over all the different objects, # returning a dictionary of <object_id: force_vector> f = self.obj1_ftc.read(without_bias = not bias) f = f[0:3, :] trans, quat = self.tf_lstnr.lookupTransform('/torso_lift_link', '/ft2', rospy.Time(0)) rot = tr.quaternion_to_matrix(quat) f = rot * f return -f # the negative is intentional (Advait, Nov 24. 2010.) def bias_fts(self): self.obj1_ftc.bias() def get_arrow_text_markers(p, f, frame, m_id, duration): t_now = rospy.Time.now() q = hv.arrow_direction_to_quat(f) arrow_len = np.linalg.norm(f) * 0.04 scale = (arrow_len, 0.2, 0.2) m1 = hv.single_marker(p, q, 'arrow', frame, scale, m_id = m_id, duration = duration) m1.header.stamp = t_now m2 = hv.single_marker(p, q, 'text_view_facing', frame, (0.1, 0.1, 0.1), m_id = m_id+1, duration = duration, color=(1.,0.,0.,1.)) m2.text = '%.1f N'%(np.linalg.norm(f)) m2.header.stamp = t_now return m1, m2 if __name__ == '__main__': rospy.init_node('force_visualize_test') marker_pub = rospy.Publisher('/skin/viz_marker', Marker) fts = object_ft_sensors() fts.bias_fts() pr2_arms = pa.PR2Arms() r_arm, l_arm = 0, 1 arm = r_arm while not rospy.is_shutdown(): f = fts.get_forces() p, r = pr2_arms.end_effector_pos(arm) m1, m2 = get_arrow_text_markers(p, f, 'torso_lift_link', 0, 1.) marker_pub.publish(m1) marker_pub.publish(m2) rospy.sleep(0.1)
[ [ 1, 0, 0.0225, 0.0112, 0, 0.66, 0, 509, 0, 1, 0, 0, 509, 0, 0 ], [ 1, 0, 0.0337, 0.0112, 0, 0.66, 0.0556, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0562, 0.0112, 0, ...
[ "import sys", "import numpy as np, math", "import add_cylinder as ac", "import online_collision_detection as ocd", "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import rospy", "from mapping_msgs.msg import Collisi...
import roslib; roslib.load_manifest('arm_navigation_tutorials') import sys import rospy from planning_environment_msgs.srv import GetStateValidity from planning_environment_msgs.srv import GetStateValidityRequest if __name__ == '__main__': rospy.init_node('get_state_validity_python') srv_nm = 'environment_server_right_arm/get_state_validity' rospy.wait_for_service(srv_nm) get_state_validity = rospy.ServiceProxy(srv_nm, GetStateValidity) req = GetStateValidityRequest() req.robot_state.joint_state.name = ['r_shoulder_pan_joint', 'r_shoulder_lift_joint', 'r_upper_arm_roll_joint', 'r_elbow_flex_joint', 'r_forearm_roll_joint', 'r_wrist_flex_joint', 'r_wrist_roll_joint'] req.robot_state.joint_state.position = [0.] * 7 req.robot_state.joint_state.position[0] = 0.4 req.robot_state.joint_state.position[3] = -0.4 req.robot_state.joint_state.header.stamp = rospy.Time.now() req.check_collisions = True res = get_state_validity.call(req) if res.error_code.val == res.error_code.SUCCESS: rospy.loginfo('Requested state is not in collision') else: rospy.loginfo('Requested state is in collision. Error code: %d'%(res.error_code.val))
[ [ 1, 0, 0.05, 0.025, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.05, 0.025, 0, 0.66, 0.1667, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.075, 0.025, 0, 0.66, ...
[ "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import sys", "import rospy", "from planning_environment_msgs.srv import GetStateValidity", "from planning_environment_msgs.srv import GetStateValidityRequest", "if __nam...
import sys import numpy as np, math import add_cylinder as ac import online_collision_detection as ocd import force_visualize_test as fvt import roslib; roslib.load_manifest('arm_navigation_tutorials') import rospy from mapping_msgs.msg import CollisionObject from visualization_msgs.msg import Marker import hrl_lib.viz as hv roslib.load_manifest('hrl_pr2_lib') import hrl_pr2_lib.pr2_arms as pa def contact_info_list_to_dict(cont_info_list): ci = cont_info_list[0] frm = ci.header.frame_id # print 'frame:', frm b1 = ci.contact_body_1 b2 = ci.contact_body_2 contact_dict = {} pts_list = [] for ci in cont_info_list: if frm != ci.header.frame_id: rospy.logerr('initial frame_id: %s and current frame_id: %s'%(frm, ci.header.frame_id)) b1 = ci.contact_body_1 b2 = ci.contact_body_2 two_bodies = b1 + '+' + b2 if two_bodies not in contact_dict: contact_dict[two_bodies] = [] contact_dict[two_bodies].append((ci.position.x, ci.position.y, ci.position.z)) return contact_dict def visualize_contact_dict(cd, marker_pub, fts): color_list = [(1.,0.,0.), (0.,1.,0.), (0.,0.,1.), (1.,1.,0.), (1.,0.,1.), (0.,1.,1.), (0.5,1.,0.), (0.5,0.,1.), (0.,0.5,1.) ] pts_list = [] cs_list = [] marker_list = [] for i, k in enumerate(cd.keys()): pts = np.matrix(cd[k]).T c = color_list[i] cs = np.ones((4, pts.shape[1])) cs[0,:] = c[0] cs[1,:] = c[1] cs[2,:] = c[2] pts_list.append(pts) cs_list.append(cs) print '# of contact points:', pts.shape[1] mn = np.mean(pts, 1) f = fts.get_forces() m1, m2 = fvt.get_arrow_text_markers(mn, f, 'base_footprint', m_id = 2*i+1, duration=0.5) marker_pub.publish(m1) marker_pub.publish(m2) m = hv.list_marker(np.column_stack(pts_list), np.column_stack(cs_list), (0.01, 0.01, 0.01), 'points', 'base_footprint', duration=1.0, m_id=0) t_now = rospy.Time.now() m.header.stamp = t_now marker_pub.publish(m) for m in marker_list: m.header.stamp = rospy.Time.now() marker_pub.publish(m) if __name__ == '__main__': rospy.init_node('pr2_skin_simulate') pub = rospy.Publisher('collision_object', CollisionObject) marker_pub = rospy.Publisher('/skin/viz_marker', Marker) fts = fvt.object_ft_sensors() fts.bias_fts() pr2_arms = pa.PR2Arms() r_arm, l_arm = 0, 1 arm = r_arm raw_input('Touch the object and then hit ENTER.') ee_pos, ee_rot = pr2_arms.end_effector_pos(arm) print 'ee_pos:', ee_pos.flatten() print 'ee_pos.shape:', ee_pos.shape trans, quat = pr2_arms.tf_lstnr.lookupTransform('/base_footprint', '/torso_lift_link', rospy.Time(0)) height = ee_pos[2] + trans[2] ee_pos[2] = -trans[2] ac.add_cylinder('pole', ee_pos, 0.02, height, '/torso_lift_link', pub) rospy.loginfo('Now starting the loop where I get contact locations.') col_det = ocd.online_collision_detector() while not rospy.is_shutdown(): rospy.sleep(0.1) res = col_det.check_validity(pr2_arms, arm) if res.error_code.val == res.error_code.SUCCESS: rospy.loginfo('No contact') else: contact_dict = contact_info_list_to_dict(res.contacts) print 'contact_dict.keys:', contact_dict.keys() visualize_contact_dict(contact_dict, marker_pub, fts)
[ [ 1, 0, 0.0172, 0.0086, 0, 0.66, 0, 509, 0, 1, 0, 0, 509, 0, 0 ], [ 1, 0, 0.0259, 0.0086, 0, 0.66, 0.0667, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0431, 0.0086, 0, ...
[ "import sys", "import numpy as np, math", "import add_cylinder as ac", "import online_collision_detection as ocd", "import force_visualize_test as fvt", "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import rospy",...
import roslib; roslib.load_manifest('arm_navigation_tutorials') import sys import rospy from mapping_msgs.msg import CollisionObject from mapping_msgs.msg import CollisionObjectOperation from geometric_shapes_msgs.msg import Shape from geometry_msgs.msg import Pose def add_cylinder(id, bottom, radius, height, frameid, pub): cylinder_object = CollisionObject() cylinder_object.id = id cylinder_object.operation.operation = CollisionObjectOperation.ADD cylinder_object.header.frame_id = frameid cylinder_object.header.stamp = rospy.Time.now() shp = Shape() shp.type = Shape.CYLINDER shp.dimensions = [0, 0] shp.dimensions[0] = radius shp.dimensions[1] = height pose = Pose() pose.position.x = bottom[0] pose.position.y = bottom[1] pose.position.z = bottom[2] + height/2. pose.orientation.x = 0 pose.orientation.y = 0 pose.orientation.z = 0 pose.orientation.w = 1 cylinder_object.shapes.append(shp) cylinder_object.poses.append(pose) pub.publish(cylinder_object) if __name__ == '__main__': rospy.init_node('add_cylinder_python') pub = rospy.Publisher('collision_object', CollisionObject) rospy.sleep(2.) add_cylinder('pole', (0.6, -0.6, 0.), 0.1, 0.75, 'base_link', pub) rospy.loginfo('Should have published') rospy.sleep(2.)
[ [ 1, 0, 0.04, 0.02, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.04, 0.02, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.06, 0.02, 0, 0.66, 0.22...
[ "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import sys", "import rospy", "from mapping_msgs.msg import CollisionObject", "from mapping_msgs.msg import CollisionObjectOperation", "from geometric_shapes_msgs.msg imp...
import roslib; roslib.load_manifest('arm_navigation_tutorials') import sys import rospy import hrl_lib.transforms as tr from planning_environment_msgs.srv import GetStateValidity from planning_environment_msgs.srv import GetStateValidityRequest from sensor_msgs.msg import JointState roslib.load_manifest('hrl_pr2_lib') import hrl_pr2_lib.pr2_arms as pa class online_collision_detector(): def __init__(self): srv_nm = 'environment_server_right_arm/get_state_validity' rospy.wait_for_service(srv_nm) self.state_validator = rospy.ServiceProxy(srv_nm, GetStateValidity, persistent=True) def check_validity(self, pr2_arms, arm): q = pr2_arms.get_joint_angles(arm) joint_nm_list = ['r_shoulder_pan_joint', 'r_shoulder_lift_joint', 'r_upper_arm_roll_joint', 'r_elbow_flex_joint', 'r_forearm_roll_joint', 'r_wrist_flex_joint', 'r_wrist_roll_joint'] req = GetStateValidityRequest() req.robot_state.joint_state.name = joint_nm_list req.robot_state.joint_state.position = q req.robot_state.joint_state.header.stamp = rospy.Time.now() req.check_collisions = True res = self.state_validator.call(req) return res if __name__ == '__main__': rospy.init_node('get_state_validity_python') pr2_arms = pa.PR2Arms() r_arm, l_arm = 0, 1 arm = r_arm col_det = online_collision_detector() while not rospy.is_shutdown(): rospy.sleep(0.1) res = col_det.check_validity(pr2_arms, arm) if res.error_code.val == res.error_code.SUCCESS: rospy.loginfo('Requested state is not in collision') else: rospy.loginfo('Requested state is in collision. Error code: %d'%(res.error_code.val))
[ [ 1, 0, 0.0333, 0.0167, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0333, 0.0167, 0, 0.66, 0.0909, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.05, 0.0167, 0, 0.66...
[ "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import sys", "import rospy", "import hrl_lib.transforms as tr", "from planning_environment_msgs.srv import GetStateValidity", "from planning_environment_msgs.srv import ...
# # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # Author: Advait Jain (advait@cc.gatech.edu), Healthcare Robotics Lab, Georgia Tech import roslib; roslib.load_manifest('arm_navigation_tutorials') import sys import rospy from planning_environment_msgs.srv import GetStateValidity from planning_environment_msgs.srv import GetStateValidityRequest if __name__ == '__main__': rospy.init_node('get_state_validity_python') srv_nm = 'environment_server_right_arm/get_state_validity' rospy.wait_for_service(srv_nm) get_state_validity = rospy.ServiceProxy(srv_nm, GetStateValidity) req = GetStateValidityRequest() req.robot_state.joint_state.name = ['r_shoulder_pan_joint', 'r_shoulder_lift_joint', 'r_upper_arm_roll_joint', 'r_elbow_flex_joint', 'r_forearm_roll_joint', 'r_wrist_flex_joint', 'r_wrist_roll_joint'] req.robot_state.joint_state.position = [0.] * 7 req.robot_state.joint_state.position[0] = 0.4 req.robot_state.joint_state.position[3] = -0.4 req.robot_state.joint_state.header.stamp = rospy.Time.now() req.check_collisions = True res = get_state_validity.call(req) if res.error_code.val == res.error_code.SUCCESS: rospy.loginfo('Requested state is not in collision') else: rospy.loginfo('Requested state is in collision. Error code: %d'%(res.error_code.val))
[ [ 1, 0, 0.4412, 0.0147, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.4412, 0.0147, 0, 0.66, 0.1667, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.4559, 0.0147, 0, 0....
[ "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import sys", "import rospy", "from planning_environment_msgs.srv import GetStateValidity", "from planning_environment_msgs.srv import GetStateValidityRequest", "if __nam...
# # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # Author: Advait Jain (advait@cc.gatech.edu), Healthcare Robotics Lab, Georgia Tech import roslib; roslib.load_manifest('arm_navigation_tutorials') import sys import rospy from mapping_msgs.msg import CollisionObject from mapping_msgs.msg import CollisionObjectOperation from geometric_shapes_msgs.msg import Shape from geometry_msgs.msg import Pose def add_cylinder(id, bottom, radius, height, frameid): cylinder_object = CollisionObject() cylinder_object.id = id cylinder_object.operation.operation = CollisionObjectOperation.ADD cylinder_object.header.frame_id = frameid cylinder_object.header.stamp = rospy.Time.now() shp = Shape() shp.type = Shape.CYLINDER shp.dimensions = [0, 0] shp.dimensions[0] = radius shp.dimensions[1] = height pose = Pose() pose.position.x = bottom[0] pose.position.y = bottom[1] pose.position.z = bottom[2] + height/2. pose.orientation.x = 0 pose.orientation.y = 0 pose.orientation.z = 0 pose.orientation.w = 1 cylinder_object.shapes.append(shp) cylinder_object.poses.append(pose) pub.publish(cylinder_object) if __name__ == '__main__': rospy.init_node('add_cylinder_python') pub = rospy.Publisher('collision_object', CollisionObject) rospy.sleep(2.) add_cylinder('pole', (0.6, -0.6, 0.), 0.1, 0.75, 'base_link') rospy.loginfo('Should have published') rospy.sleep(2.)
[ [ 1, 0, 0.3924, 0.0127, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.3924, 0.0127, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.4051, 0.0127, 0, 0....
[ "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import roslib; roslib.load_manifest('arm_navigation_tutorials')", "import sys", "import rospy", "from mapping_msgs.msg import CollisionObject", "from mapping_msgs.msg import CollisionObjectOperation", "from geometric_shapes_msgs.msg imp...
import PyKDL as kdl import numpy as np, math import roslib; roslib.load_manifest('pr2_arms_kdl') import rospy import hrl_lib.kdl_utils as ku class PR2_arm_kdl(): def __init__(self): self.right_chain = self.create_right_chain() fk, ik_v, ik_p = self.create_solvers(self.right_chain) self.right_fk = fk self.right_ik_v = ik_v self.right_ik_p = ik_p def create_right_chain(self): ch = kdl.Chain() self.right_arm_base_offset_from_torso_lift_link = np.matrix([0., -0.188, 0.]).T # shoulder pan ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotZ),kdl.Frame(kdl.Vector(0.1,0.,0.)))) # shoulder lift ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotY),kdl.Frame(kdl.Vector(0.,0.,0.)))) # upper arm roll ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotX),kdl.Frame(kdl.Vector(0.4,0.,0.)))) # elbox flex ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotY),kdl.Frame(kdl.Vector(0.0,0.,0.)))) # forearm roll ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotX),kdl.Frame(kdl.Vector(0.321,0.,0.)))) # wrist flex ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotY),kdl.Frame(kdl.Vector(0.,0.,0.)))) # wrist roll ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotX),kdl.Frame(kdl.Vector(0.,0.,0.)))) return ch def create_solvers(self, ch): fk = kdl.ChainFkSolverPos_recursive(ch) ik_v = kdl.ChainIkSolverVel_pinv(ch) ik_p = kdl.ChainIkSolverPos_NR(ch, fk, ik_v) return fk, ik_v, ik_p def FK_kdl(self, arm, q, link_number): if arm == 'right_arm': fk = self.right_fk endeffec_frame = kdl.Frame() kinematics_status = fk.JntToCart(q, endeffec_frame, link_number) if kinematics_status >= 0: return endeffec_frame else: rospy.loginfo('Could not compute forward kinematics.') return None else: msg = '%s arm not supported.'%(arm) rospy.logerr(msg) raise RuntimeError(msg) ## returns point in torso lift link. def FK_all(self, arm, q, link_number = 7): q = self.pr2_to_kdl(q) frame = self.FK_kdl(arm, q, link_number) pos = frame.p pos = ku.kdl_vec_to_np(pos) pos = pos + self.right_arm_base_offset_from_torso_lift_link m = frame.M rot = ku.kdl_rot_to_np(m) return pos, rot def kdl_to_pr2(self, q): if q == None: return None q_pr2 = [0] * 7 q_pr2[0] = q[0] q_pr2[1] = q[1] q_pr2[2] = q[2] q_pr2[3] = q[3] q_pr2[4] = q[4] q_pr2[5] = q[5] q_pr2[6] = q[6] return q_pr2 def pr2_to_kdl(self, q): if q == None: return None n = len(q) q_kdl = kdl.JntArray(n) for i in range(n): q_kdl[i] = q[i] return q_kdl if __name__ == '__main__': roslib.load_manifest('hrl_pr2_lib') import hrl_pr2_lib.pr2_arms as pa import hrl_lib.viz as hv from visualization_msgs.msg import Marker rospy.init_node('kdl_pr2_test') marker_pub = rospy.Publisher('/kdl_pr2_arms/viz_marker', Marker) pr2_arms = pa.PR2Arms(gripper_point=(0.,0.,0.)) pr2_kdl = PR2_arm_kdl() r_arm, l_arm = 0, 1 while not rospy.is_shutdown(): q = pr2_arms.get_joint_angles(r_arm) p, r = pr2_kdl.FK_all('right_arm', q, 7) m = hv.create_frame_marker(p, r, 0.3, 'torso_lift_link') time_stamp = rospy.Time.now() m.header.stamp = time_stamp marker_pub.publish(m) rospy.sleep(0.1)
[ [ 1, 0, 0.0085, 0.0085, 0, 0.66, 0, 198, 0, 1, 0, 0, 198, 0, 0 ], [ 1, 0, 0.0254, 0.0085, 0, 0.66, 0.1429, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0424, 0.0085, 0, ...
[ "import PyKDL as kdl", "import numpy as np, math", "import roslib; roslib.load_manifest('pr2_arms_kdl')", "import roslib; roslib.load_manifest('pr2_arms_kdl')", "import rospy", "import hrl_lib.kdl_utils as ku", "class PR2_arm_kdl():\n\n def __init__(self):\n self.right_chain = self.create_righ...
import numpy as np, math import roslib; roslib.load_manifest('point_cloud_ros') import rospy import hrl_tilting_hokuyo.display_3d_mayavi as d3m from point_cloud_ros.msg import OccupancyGrid import point_cloud_ros.occupancy_grid as pog ## convert OccupancyGrid message to the occupancy_grid_3d object. # @param to_binary - want the occupancy grid to be binarified. # @return occupancy_grid_3d object def og_msg_to_og3d(og, to_binary=True): c = np.matrix([og.center.x, og.center.y, og.center.z]).T s = np.matrix([og.grid_size.x, og.grid_size.y, og.grid_size.z]).T r = np.matrix([og.resolution.x, og.resolution.y, og.resolution.z]).T og3d = pog.occupancy_grid_3d(c, s, r, np.array(og.data), og.occupancy_threshold, to_binary = to_binary) return og3d ## convert occupancy_grid_3d object to OccupancyGrid message. # sets the frame to base_link and stamp to the current time. # @return OccupancyGrid object def og3d_to_og_msg(og3d): og = OccupancyGrid() og.center.x = og3d.center[0,0] og.center.y = og3d.center[1,0] og.center.z = og3d.center[2,0] og.grid_size.x = og3d.size[0,0] og.grid_size.y = og3d.size[1,0] og.grid_size.z = og3d.size[2,0] og.resolution.x = og3d.resolution[0,0] og.resolution.y = og3d.resolution[1,0] og.resolution.z = og3d.resolution[2,0] og.occupancy_threshold = 1 og.data = og3d.grid.flatten().tolist() og.header.frame_id = 'base_link' og.header.stamp = rospy.rostime.get_rostime() return og ## create an OccupancyGrid msg object for the purpose of setting the # grid parameters for pc_to_og node. # @param center - 3x1 np matrix. # @param size - 3x1 np matrix. # @param resolution - 3x1 np matrix. # @param occupancy_threshold - integer # @param frame_id - string. def og_param_msg(center, size, resolution, occupancy_threshold, frame_id): og = OccupancyGrid() og.center.x = center[0,0] og.center.y = center[1,0] og.center.z = center[2,0] og.grid_size.x = size[0,0] og.grid_size.y = size[1,0] og.grid_size.z = size[2,0] og.resolution.x = resolution[0,0] og.resolution.y = resolution[1,0] og.resolution.z = resolution[2,0] og.occupancy_threshold = occupancy_threshold og.header.frame_id = frame_id og.header.stamp = rospy.rostime.get_rostime() return og if __name__ == '__main__': rospy.init_node('og_sample_python') param_list = [None, False] rospy.Subscriber('occupancy_grid', OccupancyGrid, cb, param_list) rospy.logout('Ready') while not rospy.is_shutdown(): if param_list[1] == True: og3d = param_list[0] print 'grid_shape:', og3d.grid.shape pts = og3d.grid_to_points() print pts.shape break rospy.sleep(0.1) d3m.plot_points(pts) d3m.show()
[ [ 1, 0, 0.0217, 0.0109, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0435, 0.0109, 0, 0.66, 0.1, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0435, 0.0109, 0, 0.6...
[ "import numpy as np, math", "import roslib; roslib.load_manifest('point_cloud_ros')", "import roslib; roslib.load_manifest('point_cloud_ros')", "import rospy", "import hrl_tilting_hokuyo.display_3d_mayavi as d3m", "from point_cloud_ros.msg import OccupancyGrid", "import point_cloud_ros.occupancy_grid as...
#!/usr/bin/python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # ## Testing point_cloud_mapping from python ## author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import roslib; roslib.load_manifest('point_cloud_ros') import rospy from sensor_msgs.msg import PointCloud from geometry_msgs.msg import Point32 import numpy as np, math from numpy import pi import sys, os, optparse, time import copy import hrl_lib.util as ut import point_cloud_ros.point_cloud_utils as pcu from enthought.mayavi import mlab def SphereToCart(rho, theta, phi): x = rho * np.sin(phi) * np.cos(theta) y = rho * np.sin(phi) * np.sin(theta) z = rho * np.cos(phi) return (x,y,z) def generate_sphere(): pts = 4e3 theta = np.random.rand(pts) * 2*pi phi = np.random.rand(pts) * pi rho = 1*np.ones(len(theta)) x,y,z = SphereToCart(rho,theta,phi) pts = np.matrix(np.row_stack((x,y,z))) return pcu.np_points_to_ros(pts) def plot_cloud(pts): x = pts[0,:].A1 y = pts[1,:].A1 z = pts[2,:].A1 mlab.points3d(x,y,z,mode='point') mlab.show() def plot_normals(pts,normals,curvature=None): x = pts[0,:].A1 y = pts[1,:].A1 z = pts[2,:].A1 u = normals[0,:].A1 v = normals[1,:].A1 w = normals[2,:].A1 if curvature != None: #mlab.points3d(x,y,z,curvature,mode='point',scale_factor=1.0) mlab.points3d(x,y,z,curvature,mode='sphere',scale_factor=0.1,mask_points=1) mlab.colorbar() else: mlab.points3d(x,y,z,mode='point') mlab.quiver3d(x,y,z,u,v,w,mask_points=16,scale_factor=0.1) # mlab.axes() mlab.show() def downsample_cb(cloud_down): print 'downsample_cb got called.' pts = ros_pts_to_np(cloud_down.pts) x = pts[0,:].A1 y = pts[1,:].A1 z = pts[2,:].A1 mlab.points3d(x,y,z,mode='point') mlab.show() def normals_cb(normals_cloud): print 'normals_cb got called.' d = {} t0 = time.time() pts = ros_pts_to_np(normals_cloud.pts) t1 = time.time() print 'time to go from ROS point cloud to np matrx:', t1-t0 d['pts'] = pts if normals_cloud.chan[0].name != 'nx': print '################################################################################' print 'synthetic_point_clouds.normals_cloud: DANGER DANGER normals_cloud.chan[0] is NOT nx, it is:', normals_cloud.chan[0].name print 'Exiting...' print '################################################################################' sys.exit() normals_list = [] for i in range(3): normals_list.append(normals_cloud.chan[i].vals) d['normals'] = np.matrix(normals_list) d['curvature'] = normals_cloud.chan[3].vals print 'd[\'pts\'].shape:', d['pts'].shape print 'd[\'normals\'].shape:', d['normals'].shape ut.save_pickle(d, 'normals_cloud_'+ut.formatted_time()+'.pkl') if __name__ == '__main__': p = optparse.OptionParser() p.add_option('--sphere', action='store_true', dest='sphere', help='sample a sphere and publish the point cloud') p.add_option('--plot', action='store_true', dest='plot', help='plot the result') p.add_option('-f', action='store', type='string',dest='fname', default=None, help='pkl file with the normals.') p.add_option('--pc', action='store', type='string',dest='pc_fname', default=None, help='pkl file with 3xN numpy matrix (numpy point cloud).') opt, args = p.parse_args() sphere_flag = opt.sphere plot_flag = opt.plot fname = opt.fname pc_fname = opt.pc_fname if sphere_flag or pc_fname!=None: rospy.init_node('point_cloud_tester', anonymous=True) pub = rospy.Publisher("tilt_laser_cloud", PointCloud) rospy.Subscriber("cloud_normals", PointCloud, normals_cb) rospy.Subscriber("cloud_downsampled", PointCloud, downsample_cb) time.sleep(1) if sphere_flag: pc = generate_sphere() if pc_fname != None: pts = ut.load_pickle(pc_fname) print 'before np_points_to_ros' t0 = time.time() pc = pcu.np_points_to_ros(pts) t1 = time.time() print 'time to go from numpy to ros:', t1-t0 t0 = time.time() pcu.ros_pointcloud_to_np(pc) t1 = time.time() print 'time to go from ros to numpy:', t1-t0 pub.publish(pc) rospy.spin() if plot_flag: if fname == None: print 'Please give a pkl file for plotting (-f option)' print 'Exiting...' sys.exit() d = ut.load_pickle(fname) plot_normals(d['pts'],d['normals'],d['curvature'])
[ [ 1, 0, 0.1737, 0.0053, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.1737, 0.0053, 0, 0.66, 0.0556, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.1842, 0.0053, 0, 0....
[ "import roslib; roslib.load_manifest('point_cloud_ros')", "import roslib; roslib.load_manifest('point_cloud_ros')", "import rospy", "from sensor_msgs.msg import PointCloud", "from geometry_msgs.msg import Point32", "import numpy as np, math", "from numpy import pi", "import sys, os, optparse, time", ...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import sys, optparse, os import time import math, numpy as np import scipy.ndimage as ni import copy import hrl_lib.util as ut, hrl_lib.transforms as tr ## subtract occupancy grids. og1 = og1-og2 # # @param og1 - occupancy_grid_3d object. # @param og2 - occupancy_grid_3d object. # #will position og2 at an appropriate location within og1 (hopefully) #will copy points in og2 but not in og1 into og1 # # points corresponding to the gird cells whose occupancy drops to # zero will still be in grid_points_list #UNTESTED: # * subtracting grids of different sizes. # * how the rotation_z of the occupancy grid will affect things. def subtract(og1,og2): if np.all(og1.resolution==og2.resolution) == False: print 'occupancy_grid_3d.subtract: The resolution of the two grids is not the same.' print 'res1, res2:', og1.resolution.A1.tolist(), og2.resolution.A1.tolist() print 'Exiting...' sys.exit() sub_tlb = og2.tlb sub_brf = og2.brf sub_tlb_idx = np.round((sub_tlb-og1.brf)/og1.resolution) sub_brf_idx = np.round((sub_brf-og1.brf)/og1.resolution) x_s,y_s,z_s = int(sub_brf_idx[0,0]),int(sub_brf_idx[1,0]),int(sub_brf_idx[2,0]) x_e,y_e,z_e = int(sub_tlb_idx[0,0]),int(sub_tlb_idx[1,0]),int(sub_tlb_idx[2,0]) x_e = min(x_e+1,og1.grid_shape[0,0]) y_e = min(y_e+1,og1.grid_shape[1,0]) z_e = min(z_e+1,og1.grid_shape[2,0]) sub_grid = og1.grid[x_s:x_e,y_s:y_e,z_s:z_e] if np.any(og1.grid_shape!=og2.grid_shape): print '#############################################################################' print 'WARNING: occupancy_grid_3d.subtract has not been tested for grids of different sizes.' print '#############################################################################' sub_grid = sub_grid-og2.grid sub_grid = np.abs(sub_grid) # for now. og1.grid[x_s:x_e,y_s:y_e,z_s:z_e] = sub_grid idxs = np.where(sub_grid>=1) shp = og2.grid_shape list_idxs = (idxs[0]+idxs[1]*shp[0,0]+idxs[2]*shp[0,0]*shp[1,0]).tolist() og1_list_idxs = (idxs[0]+x_s+(idxs[1]+y_s)*shp[0,0]+(idxs[2]+z_s)*shp[0,0]*shp[1,0]).tolist() og1_list_len = len(og1.grid_points_list) for og1_pts_idxs,pts_idxs in zip(og1_list_idxs,list_idxs): if og1_pts_idxs<og1_list_len: og1.grid_points_list[og1_pts_idxs] += og2.grid_points_list[pts_idxs] ## class which implements the occupancy grid class occupancy_grid_3d(): ## # @param brf - 3x1 matrix. Bottom Right Front. # @param tlb - 3x1 matrix (coord of center of the top left back cell) # @param resolution - 3x1 matrix. size of each cell (in meters) along # the different directions. def __init__(self, brf, tlb, resolution, rotation_z=math.radians(0.)): #print np.round((tlb-brf)/resolution).astype('int')+1 self.grid = np.zeros(np.round((tlb-brf)/resolution).astype('int')+1,dtype='int') self.tlb = tlb self.brf = brf self.grid_shape = np.matrix(self.grid.shape).T self.resolution = resolution n_cells = self.grid.shape[0]*self.grid.shape[1]*self.grid.shape[2] self.grid_points_list = [[] for i in range(n_cells)] self.rotation_z = rotation_z ## returns list of 8 tuples of 3x1 points which form the edges of the grid. # Useful for displaying the extents of the volume of interest (VOI). # @return list of 8 tuples of 3x1 points which form the edges of the grid. def grid_lines(self, rotation_angle=0.): grid_size = np.multiply(self.grid_shape,self.resolution) rot_mat = tr.rotZ(rotation_angle) p5 = self.tlb p6 = p5+np.matrix([0.,-grid_size[1,0],0.]).T p8 = p5+np.matrix([0.,0.,-grid_size[2,0]]).T p7 = p8+np.matrix([0.,-grid_size[1,0],0.]).T p3 = self.brf p4 = p3+np.matrix([0.,grid_size[1,0],0.]).T p2 = p3+np.matrix([0.,0.,grid_size[2,0]]).T p1 = p2+np.matrix([0.,grid_size[1,0],0.]).T p1 = rot_mat*p1 p2 = rot_mat*p2 p3 = rot_mat*p3 p4 = rot_mat*p4 p5 = rot_mat*p5 p6 = rot_mat*p6 p7 = rot_mat*p7 p8 = rot_mat*p8 l = [(p1,p2),(p1,p4),(p2,p3),(p3,p4),(p5,p6),(p6,p7),(p7,p8),(p8,p5),(p1,p5),(p2,p6),(p4,p8),(p3,p7)] #l = [(p5,p6),(p5,p3),(p1,p2)] return l ## fill the occupancy grid. # @param pts - 3xN matrix of points. # @param ignore_z - not use the z coord of the points. grid will be like a 2D grid. # #each cell of the grid gets filled the number of points that fall in the cell. def fill_grid(self,pts,ignore_z=False): if ignore_z: idx = np.where(np.min(np.multiply(pts[0:2,:]>self.brf[0:2,:], pts[0:2,:]<self.tlb[0:2,:]),0))[1] else: idx = np.where(np.min(np.multiply(pts[0:3,:]>self.brf,pts[0:3,:]<self.tlb),0))[1] if idx.shape[1] == 0: print 'aha!' return pts = pts[:,idx.A1.tolist()] # Find coordinates p_all = np.round((pts[0:3,:]-self.brf)/self.resolution) # Rotate points pts[0:3,:] = tr.Rz(self.rotation_z).T*pts[0:3,:] for i,p in enumerate(p_all.astype('int').T): if ignore_z: p[0,2] = 0 if np.any(p<0) or np.any(p>=self.grid_shape.T): continue tup = tuple(p.A1) self.grid_points_list[ tup[0] + self.grid_shape[0,0] * tup[1] + self.grid_shape[0,0] * self.grid_shape[1,0] * tup[2]].append(pts[:,i]) self.grid[tuple(p.A1)] += 1 def to_binary(self,thresh=1): ''' all cells with occupancy>=thresh set to 1, others set to 0. ''' filled = (self.grid>=thresh) self.grid[np.where(filled==True)] = 1 self.grid[np.where(filled==False)] = 0 def argmax_z(self,index_min=-np.Inf,index_max=np.Inf,search_up=False,search_down=False): ''' searches in the z direction for maximum number of cells with occupancy==1 call this function after calling to_binary() returns index. ''' index_min = int(max(index_min,0)) index_max = int(min(index_max,self.grid_shape[2,0]-1)) z_count_mat = [] #for i in xrange(self.grid_shape[2,0]): for i in xrange(index_min,index_max+1): z_count_mat.append(np.where(self.grid[:,:,i]==1)[0].shape[0]) if z_count_mat == []: return None z_count_mat = np.matrix(z_count_mat).T max_z = np.argmax(z_count_mat) max_count = z_count_mat[max_z,0] max_z += index_min print '#### max_count:', max_count if search_up: max_z_temp = max_z for i in range(1,5): #if (z_count_mat[max_z+i,0]*3.0)>max_count: #A #if (z_count_mat[max_z+i,0]*8.0)>max_count: #B if (max_z+i)>index_max: break if (z_count_mat[max_z+i-index_min,0]*5.0)>max_count: #B' max_z_temp = max_z+i max_z = max_z_temp if search_down: max_z_temp = max_z for i in range(1,5): if (max_z-i)<index_min: break if (max_z-i)>index_max: continue if (z_count_mat[max_z-i-index_min,0]*5.0)>max_count: max_z_temp = max_z-i max_z = max_z_temp return max_z,max_count def find_plane_indices(self,hmin=-np.Inf,hmax=np.Inf,assume_plane=False): ''' assume_plane - always return something. returns list of indices (z) corrresponding to horizontal plane points. returns [] if there is no plane ''' index_min = int(max(round((hmin-self.brf[2,0])/self.resolution[2,0]),0)) index_max = int(min(round((hmax-self.brf[2,0])/self.resolution[2,0]),self.grid_shape[2,0]-1)) z_plane,max_count = self.argmax_z(index_min,index_max,search_up=True) if z_plane == None: print 'oink oink.' return [] #---------- A # extra_remove_meters = 0.01 # n_more_to_remove = int(round(extra_remove_meters/self.resolution[2,0])) # l = range(max(z_plane-n_more_to_remove-1,0), # min(z_plane+n_more_to_remove+1,self.grid_shape[2,0]-1)) #---------- B extra_remove_meters = 0.005 n_more_to_remove = int(round(extra_remove_meters/self.resolution[2,0])) l = range(max(z_plane-10,0), min(z_plane+n_more_to_remove+1,self.grid_shape[2,0]-1)) # figure out whether this is indeed a plane. if assume_plane == False: n_more = int(round(0.1/self.resolution[2,0])) l_confirm = l+ range(max(l),min(z_plane+n_more+1,self.grid_shape[2,0]-1)) grid_2d = np.max(self.grid[:,:,l],2) n_plane_cells = grid_2d.sum() grid_2d = ni.binary_fill_holes(grid_2d) # I want 4-connectivity while filling holes. n_plane_cells = grid_2d.sum() min_plane_pts_threshold = (self.grid_shape[0,0]*self.grid_shape[1,0])/4 print '###n_plane_cells:', n_plane_cells print 'min_plane_pts_threshold:', min_plane_pts_threshold print 'find_plane_indices grid shape:',self.grid_shape.T if n_plane_cells < min_plane_pts_threshold: print 'occupancy_grid_3d.find_plane_indices: There is no plane.' print 'n_plane_cells:', n_plane_cells print 'min_plane_pts_threshold:', min_plane_pts_threshold l = [] return l ## get centroids of all the occupied cells as a 3xN np matrix # @param occupancy_threshold - number of points in a cell for it to be "occupied" # @return 3xN matrix of 3d coord of the cells which have occupancy >= occupancy_threshold def grid_to_centroids(self,occupancy_threshold=1): p = np.matrix(np.row_stack(np.where(self.grid>=occupancy_threshold))).astype('float') p[0,:] = p[0,:]*self.resolution[0,0] p[1,:] = p[1,:]*self.resolution[1,0] p[2,:] = p[2,:]*self.resolution[2,0] p += self.brf return p def grid_to_points(self,array=None,occupancy_threshold=1): ''' array - if not None then this will be used instead of self.grid returns 3xN matrix of 3d coord of the cells which have occupancy >= occupancy_threshold ''' if array == None: array = self.grid idxs = np.where(array>=occupancy_threshold) list_idxs = (idxs[0]+idxs[1]*self.grid_shape[0,0]+idxs[2]*self.grid_shape[0,0]*self.grid_shape[1,0]).tolist() l = [] for pts_idxs in list_idxs: l += self.grid_points_list[pts_idxs] if l == []: p = np.matrix([]) else: p = np.column_stack(l) return p def labeled_array_to_points(self,array,label): ''' returns coordinates of centers of grid cells corresponding to label as a 3xN matrix. ''' idxs = np.where(array==label) list_idxs = (idxs[0]+idxs[1]*self.grid_shape[0,0]+idxs[2]*self.grid_shape[0,0]*self.grid_shape[1,0]).tolist() l = [] for pts_idxs in list_idxs: l += self.grid_points_list[pts_idxs] if l == []: p = np.matrix([]) else: p = np.column_stack(l) return p def remove_vertical_plane(self): ''' removes plane parallel to the YZ plane. changes grid. returns plane_indices, slice corresponding to the vertical plane. points behind the plane are lost for ever! ''' self.grid = self.grid.swapaxes(2,0) self.grid_shape = np.matrix(self.grid.shape).T # z_max_first,max_count = self.argmax_z(search_up=False) # z_max_second,max_count_second = self.argmax_z(index_min=z_max_first+int(round(0.03/self.resolution[0,0])) ,search_up=False) z_max_first,max_count = self.argmax_z(search_down=False) z_max_second,max_count_second = self.argmax_z(index_min=z_max_first+int(round(0.035/self.resolution[0,0])) ,search_down=False) z_max_first,max_count = self.argmax_z(search_down=False) #z_max = self.argmax_z(search_up=True) if (max_count_second*1./max_count) > 0.3: z_max = z_max_second else: z_max = z_max_first print 'z_max_first', z_max_first print 'z_max_second', z_max_second print 'z_max', z_max more = int(round(0.03/self.resolution[0,0])) plane_indices = range(max(0,z_max-more),min(z_max+more,self.grid_shape[2,0])) self.grid = self.grid.swapaxes(2,0) self.grid_shape = np.matrix(self.grid.shape).T ver_plane_slice = self.grid[plane_indices,:,:] self.grid[plane_indices,:,:] = 0 max_x = max(plane_indices) behind_indices = range(max_x,self.grid_shape[0,0]) self.grid[behind_indices,:,:] = 0 return plane_indices,ver_plane_slice def remove_horizontal_plane(self, remove_below=True,hmin=-np.Inf,hmax=np.Inf, extra_layers=0): ''' call after to_binary() removes points corresponding to the horizontal plane from the grid. remove_below - remove points below the plane also. hmin,hmax - min and max possible height of the plane. (meters) This function changes grid. extra_layers - number of layers above the plane to remove. Sometimes I want to be over zealous while removing plane points. e.g. max_fwd_without_collision it returns the slice which has been set to zero, in case you want to leave the grid unchanged. ''' l = self.find_plane_indices(hmin,hmax) if l == []: print 'occupancy_grid_3d.remove_horizontal_plane: No plane found.' return None,l add_num = min(10,self.grid_shape[2,0]-max(l)-1) max_l = max(l)+add_num l_edge = l+range(max(l),max_l+1) grid_2d = np.max(self.grid[:,:,l_edge],2) # grid_2d = ni.binary_dilation(grid_2d,iterations=1) # I want 4-connectivity while filling holes. grid_2d = ni.binary_fill_holes(grid_2d) # I want 4-connectivity while filling holes. connect_structure = np.empty((3,3),dtype='int') connect_structure[:,:] = 1 eroded_2d = ni.binary_erosion(grid_2d,connect_structure,iterations=2) grid_2d = grid_2d-eroded_2d idxs = np.where(grid_2d!=0) if max_l>max(l): for i in range(min(5,add_num)): self.grid[idxs[0],idxs[1],max(l)+i+1] = 0 if remove_below: l = range(0,min(l)+1)+l max_z = max(l) for i in range(extra_layers): l.append(max_z+i+1) l_edge = l+range(max(l),max_l+1) plane_and_below_pts = self.grid[:,:,l_edge] self.grid[:,:,l] = 0 # set occupancy to zero. return plane_and_below_pts,l_edge def segment_objects(self, twod=False): ''' segments out objects after removing the plane. call after calling to_binary. returns labelled_array,n_labels labelled_array - same dimen as occupancy grid, each object has a different label. ''' plane_and_below_pts,l = self.remove_horizontal_plane(extra_layers=0) if l == []: print 'occupancy_grid_3d.segment_objects: There is no plane.' return None,None if twod == False: labelled_arr,n_labels = self.find_objects() else: labelled_arr,n_labels = self.find_objects_2d() self.grid[:,:,l] = plane_and_below_pts return labelled_arr,n_labels def find_objects_2d(self): ''' projects all points into the xy plane and then performs segmentation by region growing. ''' connect_structure = np.empty((3,3),dtype='int') connect_structure[:,:] = 1 grid_2d = np.max(self.grid[:,:,:],2) # grid_2d = ni.binary_erosion(grid_2d) # grid_2d = ni.binary_erosion(grid_2d,connect_structure) labeled_arr,n_labels = ni.label(grid_2d,connect_structure) print 'found %d objects'%(n_labels) labeled_arr_3d = self.grid.swapaxes(2,0) labeled_arr_3d = labeled_arr_3d.swapaxes(1,2) print 'labeled_arr.shape:',labeled_arr.shape print 'labeled_arr_3d.shape:',labeled_arr_3d.shape labeled_arr_3d = labeled_arr_3d*labeled_arr labeled_arr_3d = labeled_arr_3d.swapaxes(2,0) labeled_arr_3d = labeled_arr_3d.swapaxes(1,0) labeled_arr = labeled_arr_3d # I still want to count cells in 3d (thin but tall objects.) if n_labels > 0: labels_list = range(1,n_labels+1) #count_objects = ni.sum(grid_2d,labeled_arr,labels_list) count_objects = ni.sum(self.grid,labeled_arr,labels_list) if n_labels == 1: count_objects = [count_objects] t0 = time.time() new_labels_list = [] for c,l in zip(count_objects,labels_list): if c > 3: new_labels_list.append(l) else: labeled_arr[np.where(labeled_arr == l)] = 0 # relabel stuff for nl,l in enumerate(new_labels_list): labeled_arr[np.where(labeled_arr == l)] = nl+1 n_labels = len(new_labels_list) t1 = time.time() print 'time:', t1-t0 print 'found %d objects'%(n_labels) # return labeled_arr,n_labels return labeled_arr_3d,n_labels def find_objects(self): ''' region growing kind of thing for segmentation. Useful if plane has been removed. ''' connect_structure = np.empty((3,3,3),dtype='int') grid = copy.copy(self.grid) connect_structure[:,:,:] = 0 connect_structure[1,1,:] = 1 iterations = int(round(0.005/self.resolution[2,0])) # iterations=5 #grid = ni.binary_closing(grid,connect_structure,iterations=iterations) connect_structure[:,:,:] = 1 labeled_arr,n_labels = ni.label(grid,connect_structure) print 'ho!' print 'found %d objects'%(n_labels) if n_labels == 0: return labeled_arr,n_labels labels_list = range(1,n_labels+1) count_objects = ni.sum(grid,labeled_arr,labels_list) if n_labels == 1: count_objects = [count_objects] # t0 = time.time() # remove_labels = np.where(np.matrix(count_objects) <= 5)[1].A1.tolist() # for r in remove_labels: # labeled_arr[np.where(labeled_arr == r)] = 0 # t1 = time.time() # labeled_arr,n_labels = ni.label(labeled_arr,connect_structure) # print 'time:', t1-t0 t0 = time.time() new_labels_list = [] for c,l in zip(count_objects,labels_list): if c > 3: new_labels_list.append(l) else: labeled_arr[np.where(labeled_arr == l)] = 0 # relabel stuff for nl,l in enumerate(new_labels_list): labeled_arr[np.where(labeled_arr == l)] = nl+1 n_labels = len(new_labels_list) t1 = time.time() print 'time:', t1-t0 print 'found %d objects'%(n_labels) return labeled_arr,n_labels if __name__ == '__main__': import pygame_opengl_3d_display as po3d import hokuyo.pygame_utils as pu import processing_3d as p3d p = optparse.OptionParser() p.add_option('-f', action='store', type='string', dest='pkl_file_name', help='file.pkl File with the scan,pos dict.',default=None) p.add_option('-c', action='store', type='string', dest='pts_pkl', help='pkl file with 3D points',default=None) opt, args = p.parse_args() pts_pkl = opt.pts_pkl pkl_file_name = opt.pkl_file_name #-------------- simple test --------------- # gr = occupancy_grid_3d(np.matrix([0.,0.,0]).T, np.matrix([1.,1.,1]).T, # np.matrix([1,1,1]).T) # pts = np.matrix([[1.1,0,-0.2],[0,0,0],[0.7,0.7,0.3],[0.6,0.8,-0.2]]).T # gr.fill_grid(pts) ## print gr.grid resolution = np.matrix([0.01,0.01,0.01]).T gr = occupancy_grid_3d(np.matrix([0.45,-0.5,-1.0]).T, np.matrix([0.65,0.05,-0.2]).T, resolution) if pts_pkl != None: pts = ut.load_pickle(pts_pkl) elif pkl_file_name != None: dict = ut.load_pickle(pkl_file_name) pos_list = dict['pos_list'] scan_list = dict['scan_list'] min_angle = math.radians(-40) max_angle = math.radians(40) l1 = dict['l1'] l2 = dict['l2'] pts = p3d.generate_pointcloud(pos_list, scan_list, min_angle, max_angle, l1, l2) else: print 'specify a pkl file -c or -f' print 'Exiting...' sys.exit() print 'started filling the grid' t0 = time.time() gr.fill_grid(pts) t1 = time.time() print 'time to fill the grid:', t1-t0 #grid_pts = gr.grid_to_points() grid_pts = gr.grid_to_centroids() ## print grid_pts cloud = pu.CubeCloud(grid_pts,(0,0,0),(resolution/2).A1.tolist()) pc = pu.PointCloud(pts,(100,100,100)) lc = pu.LineCloud(gr.grid_lines(),(100,100,0)) po3d.run([cloud,pc,lc])
[ [ 1, 0, 0.0489, 0.0016, 0, 0.66, 0, 509, 0, 3, 0, 0, 509, 0, 0 ], [ 1, 0, 0.0506, 0.0016, 0, 0.66, 0.125, 654, 0, 1, 0, 0, 654, 0, 0 ], [ 1, 0, 0.0538, 0.0016, 0, 0...
[ "import sys, optparse, os", "import time", "import math, numpy as np", "import scipy.ndimage as ni", "import copy", "import hrl_lib.util as ut, hrl_lib.transforms as tr", "def subtract(og1,og2):\n\n if np.all(og1.resolution==og2.resolution) == False:\n print('occupancy_grid_3d.subtract: The re...
import roslib; roslib.load_manifest('point_cloud_ros') from sensor_msgs.msg import PointCloud from geometry_msgs.msg import Point32 from sensor_msgs.msg import ChannelFloat32 import numpy as np import time ## PointCloud -> 3xN np matrix # @param ros_pointcloud - robot_msgs/PointCloud # @return 3xN np matrix def ros_pointcloud_to_np(ros_pointcloud): ''' ros PointCloud.pts -> 3xN numpy matrix ''' return ros_pts_to_np(ros_pointcloud.points) ## list of Point32 points -> 3xN np matrix # @param ros_points - Point32[ ] (for e.g. from robot_msgs/PointCloud or Polygon3D) # @return 3xN np matrix def ros_pts_to_np(ros_pts): pts_list = [] for p in ros_pts: pts_list.append([p.x,p.y,p.z]) return np.matrix(pts_list).T ## 3xN np matrix -> ros PointCloud # @param pts - 3xN np matrix # @return PointCloud as defined in robot_msgs/msg/PointCloud.msg def np_points_to_ros(pts): p_list = [] chlist = [] # p_list = [Point32(p[0,0], p[0,1], p[0,2]) for p in pts.T] # chlist = np.zeros(pts.shape[1]).tolist() for p in pts.T: p_list.append(Point32(p[0,0],p[0,1],p[0,2])) chlist.append(0.) ch = ChannelFloat32('t',chlist) pc = PointCloud() pc.points = p_list pc.channels = [ch] return pc
[ [ 1, 0, 0.0222, 0.0222, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0222, 0.0222, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0444, 0.0222, 0, 0....
[ "import roslib; roslib.load_manifest('point_cloud_ros')", "import roslib; roslib.load_manifest('point_cloud_ros')", "from sensor_msgs.msg import PointCloud", "from geometry_msgs.msg import Point32", "from sensor_msgs.msg import ChannelFloat32", "import numpy as np", "import time", "def ros_pointcloud_...
#!/usr/bin/python import numpy as np, math import scipy.ndimage as ni import roslib; roslib.load_manifest('point_cloud_ros') import rospy import hrl_lib.util as ut import point_cloud_ros.occupancy_grid as pog import point_cloud_ros.ros_occupancy_grid as rog from point_cloud_ros.msg import OccupancyGrid def og_cb(og_msg, param_list): global occupancy_difference_threshold, connected_comonents_size_threshold rospy.loginfo('og_cb called') diff_og = param_list[0] curr_og = rog.og_msg_to_og3d(og_msg, to_binary = False) if diff_og == None: param_list[0] = curr_og return pog.subtract(diff_og, curr_og) param_list[0] = curr_og diff_og.to_binary(occupancy_difference_threshold) # filter the noise connect_structure = np.zeros((3,3,3), dtype=int) connect_structure[1,1,:] = 1 # connect_structure[1,1,0] = 0 diff_og.grid = ni.binary_opening(diff_og.grid, connect_structure, iterations = 1) # diff_og.grid, n_labels = diff_og.connected_comonents(connected_comonents_size_threshold) print 'np.all(diff_og == 0)', np.all(diff_og.grid == 0) diff_og_msg = rog.og3d_to_og_msg(diff_og) diff_og_msg.header.frame_id = og_msg.header.frame_id diff_og_msg.header.stamp = og_msg.header.stamp param_list[1].publish(diff_og_msg) #------ arbitrarily set paramters ------- occupancy_difference_threshold = 5 connected_comonents_size_threshold = 10 if __name__ == '__main__': rospy.init_node('pc_difference_node') pub = rospy.Publisher('difference_occupancy_grid', OccupancyGrid) param_list = [None, pub] rospy.Subscriber('occupancy_grid', OccupancyGrid, og_cb, param_list) rospy.logout('Ready') rospy.spin()
[ [ 1, 0, 0.0508, 0.0169, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0678, 0.0169, 0, 0.66, 0.0833, 348, 0, 1, 0, 0, 348, 0, 0 ], [ 1, 0, 0.1017, 0.0169, 0, ...
[ "import numpy as np, math", "import scipy.ndimage as ni", "import roslib; roslib.load_manifest('point_cloud_ros')", "import roslib; roslib.load_manifest('point_cloud_ros')", "import rospy", "import hrl_lib.util as ut", "import point_cloud_ros.occupancy_grid as pog", "import point_cloud_ros.ros_occupan...
import time import sys, os, copy import numpy as np, math import scipy.ndimage as ni class occupancy_grid_3d(): ## # @param resolution - 3x1 matrix. size of each cell (in meters) along # the different directions. def __init__(self, center, size, resolution, data, occupancy_threshold, to_binary = True): self.grid_shape = size/resolution tlb = center + size/2 brf = center + size/2 self.size = size self.grid = np.reshape(data, self.grid_shape) self.grid_shape = np.matrix(self.grid.shape).T self.resolution = resolution self.center = center if to_binary: self.to_binary(occupancy_threshold) ## binarize the grid # @param occupancy_threshold - voxels with occupancy less than this are set to zero. def to_binary(self, occupancy_threshold): filled = (self.grid >= occupancy_threshold) self.grid[np.where(filled==True)] = 1 self.grid[np.where(filled==False)] = 0 ## # @param array - if not None then this will be used instead of self.grid # @return 3xN matrix of 3d coord of the cells which have occupancy = 1 def grid_to_points(self, array=None): if array == None: array = self.grid idxs = np.where(array == 1) x_idx = idxs[0] y_idx = idxs[1] z_idx = idxs[2] x = x_idx * self.resolution[0,0] + self.center[0,0] - self.size[0,0]/2 y = y_idx * self.resolution[1,0] + self.center[1,0] - self.size[1,0]/2 z = z_idx * self.resolution[2,0] + self.center[2,0] - self.size[2,0]/2 return np.matrix(np.row_stack([x,y,z])) ## 27-connected components. # @param threshold - min allowed size of connected component def connected_comonents(self, threshold): connect_structure = np.ones((3,3,3), dtype='int') grid = self.grid labeled_arr, n_labels = ni.label(grid, connect_structure) if n_labels == 0: return labeled_arr, n_labels labels_list = range(1,n_labels+1) count_objects = ni.sum(grid, labeled_arr, labels_list) if n_labels == 1: count_objects = [count_objects] t0 = time.time() new_labels_list = [] for c,l in zip(count_objects, labels_list): if c > threshold: new_labels_list.append(l) else: labeled_arr[np.where(labeled_arr == l)] = 0 # relabel stuff for nl,l in enumerate(new_labels_list): labeled_arr[np.where(labeled_arr == l)] = nl+1 n_labels = len(new_labels_list) t1 = time.time() print 'time:', t1-t0 return labeled_arr, n_labels ## subtract occupancy grids. og1 = abs(og1-og2) # @param og1 - occupancy_grid_3d object. # @param og2 - occupancy_grid_3d object. # # will position og2 at an appropriate location within og1 (hopefully) # will copy points in og2 but not in og1 into og1 # #UNTESTED: # * subtracting grids of different sizes. def subtract(og1, og2): if np.all(og1.resolution == og2.resolution) == False: print 'occupancy_grid_3d.subtract: The resolution of the two grids is not the same.' print 'res1, res2:', og1.resolution.A1.tolist(), og2.resolution.A1.tolist() return if np.any(og1.grid_shape!=og2.grid_shape): print 'Grid Sizes:', og1.grid_shape.A1, og2.grid_shape.A1 raise RuntimeError('grids are of different sizes') og1.grid = np.abs(og1.grid - og2.grid) if __name__ == '__main__': print 'Hello World'
[ [ 1, 0, 0.0183, 0.0092, 0, 0.66, 0, 654, 0, 1, 0, 0, 654, 0, 0 ], [ 1, 0, 0.0275, 0.0092, 0, 0.66, 0.1667, 509, 0, 3, 0, 0, 509, 0, 0 ], [ 1, 0, 0.0367, 0.0092, 0, ...
[ "import time", "import sys, os, copy", "import numpy as np, math", "import scipy.ndimage as ni", "class occupancy_grid_3d():\n\n ##\n # @param resolution - 3x1 matrix. size of each cell (in meters) along\n # the different directions.\n def __init__(self, center, size, resolut...
#!/usr/bin/python import numpy as np, math import time import roslib; roslib.load_manifest('point_cloud_ros') import rospy from point_cloud_ros.msg import OccupancyGrid import hrl_tilting_hokuyo.display_3d_mayavi as d3m import point_cloud_ros.ros_occupancy_grid as rog def mayavi_cb(og, param_list): og3d = rog.og_msg_to_og3d(og) param_list[0] = og3d param_list[1] = True def relay_cb(og, og_pub): rospy.logout('relay_cb called') og3d = rog.og_msg_to_og3d(og) og_new = rog.og3d_to_og_msg(og3d) og_pub.publish(og_new) def vis_occupancy_cb(og, param_list): og3d = rog.og_msg_to_og3d(og, to_binary = False) param_list[0] = og3d param_list[1] = True if __name__ == '__main__': og_pub = rospy.Publisher('relay_og_out', OccupancyGrid) rospy.Subscriber('relay_og_in', OccupancyGrid, relay_cb, og_pub) param_list = [None, False] rospy.init_node('og_sample_python') rospy.logout('Ready') #mode = rospy.get_param('~mode') #mode = 'mayavi' mode = 'vis_occupancy' if mode == 'mayavi': rospy.Subscriber('occupancy_grid', OccupancyGrid, mayavi_cb, param_list) while not rospy.is_shutdown(): if param_list[1] == True: og3d = param_list[0] print 'grid_shape:', og3d.grid.shape pts = og3d.grid_to_points() print pts.shape # param_list[1] = False break rospy.sleep(0.1) d3m.plot_points(pts) d3m.show() if mode == 'vis_occupancy': rospy.Subscriber('occupancy_grid', OccupancyGrid, vis_occupancy_cb, param_list) import matplotlib_util.util as mpu while not rospy.is_shutdown(): if param_list[1] == True: og3d = param_list[0] break rospy.sleep(0.1) occ_array = og3d.grid.flatten() mpu.pl.hist(occ_array, 100) # mpu.plot_yx(occ_array) mpu.show() elif mode == 'relay': rospy.spin()
[ [ 1, 0, 0.0375, 0.0125, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.05, 0.0125, 0, 0.66, 0.0909, 654, 0, 1, 0, 0, 654, 0, 0 ], [ 1, 0, 0.075, 0.0125, 0, 0.6...
[ "import numpy as np, math", "import time", "import roslib; roslib.load_manifest('point_cloud_ros')", "import roslib; roslib.load_manifest('point_cloud_ros')", "import rospy", "from point_cloud_ros.msg import OccupancyGrid", "import hrl_tilting_hokuyo.display_3d_mayavi as d3m", "import point_cloud_ros....
#!/usr/bin/python import numpy as np, math import roslib; roslib.load_manifest('point_cloud_ros') import rospy import point_cloud_ros.ros_occupancy_grid as rog from point_cloud_ros.msg import OccupancyGrid from visualization_msgs.msg import Marker ## return a Marker message object # @param loc - (x,y,z) # @param scale - (x,y,z) # @param color - (r,g,b,a) # @param shape - 'sphere' # @param frame_id - string. def simple_viz_marker(loc, scale, color, shape, frame_id): marker = Marker() marker.header.frame_id = frame_id marker.header.stamp = rospy.rostime.get_rostime() marker.ns = 'basic_shapes' marker.id = 0 if shape == 'sphere': marker.type = Marker.SPHERE marker.action = Marker.ADD marker.pose.position.x = loc[0] marker.pose.position.y = loc[1] marker.pose.position.z = loc[2] marker.pose.orientation.x = 0.0 marker.pose.orientation.y = 0.0 marker.pose.orientation.z = 0.0 marker.pose.orientation.w = 1.0 marker.scale.x = scale[0] marker.scale.y = scale[1] marker.scale.z = scale[2] marker.color.r = color[0] marker.color.g = color[1] marker.color.b = color[2] marker.color.a = color[3] marker.lifetime = rospy.Duration() return marker if __name__ == '__main__': rospy.init_node('set_og_param_node') og_param_pub = rospy.Publisher('/og_params', OccupancyGrid) marker_pub = rospy.Publisher('/occupancy_grid_viz_marker', Marker) rospy.logout('Ready') center = np.matrix([0.8, 0., 0.8]).T # for single object bag file #center = np.matrix([0.6, 0., 0.75]).T size = np.matrix([0.4, 0.4, 0.4]).T #resolution = np.matrix([0.01, 0.01, 0.01]).T resolution = np.matrix([0.005, 0.005, 0.005]).T occupancy_threshold = 5 frame_id = 'base_link' scale = (0.02, 0.02, 0.02) color = (0., 1., 0., 1.) shape = 'sphere' marker = simple_viz_marker(center.A1, scale, color, shape, frame_id) og_param = rog.og_param_msg(center, size, resolution, occupancy_threshold, frame_id) r = rospy.Rate(2) while not rospy.is_shutdown(): marker_pub.publish(marker) og_param_pub.publish(og_param) r.sleep() rospy.spin()
[ [ 1, 0, 0.04, 0.0133, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0667, 0.0133, 0, 0.66, 0.125, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0667, 0.0133, 0, 0.6...
[ "import numpy as np, math", "import roslib; roslib.load_manifest('point_cloud_ros')", "import roslib; roslib.load_manifest('point_cloud_ros')", "import rospy", "import point_cloud_ros.ros_occupancy_grid as rog", "from point_cloud_ros.msg import OccupancyGrid", "from visualization_msgs.msg import Marker"...
# # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # # Author: Advait Jain (advait@cc.gatech.edu), Healthcare Robotics Lab, Georgia Tech import roslib; roslib.load_manifest('move_arm_tutorials') import rospy import actionlib import geometric_shapes_msgs from move_arm_msgs.msg import MoveArmAction from move_arm_msgs.msg import MoveArmGoal from motion_planning_msgs.msg import SimplePoseConstraint from motion_planning_msgs.msg import PositionConstraint from motion_planning_msgs.msg import OrientationConstraint from actionlib_msgs.msg import GoalStatus def pose_constraint_to_position_orientation_constraints(pose_constraint): position_constraint = PositionConstraint() orientation_constraint = OrientationConstraint() position_constraint.header = pose_constraint.header position_constraint.link_name = pose_constraint.link_name position_constraint.position = pose_constraint.pose.position position_constraint.constraint_region_shape.type = geometric_shapes_msgs.msg.Shape.BOX position_constraint.constraint_region_shape.dimensions.append(2*pose_constraint.absolute_position_tolerance.x) position_constraint.constraint_region_shape.dimensions.append(2*pose_constraint.absolute_position_tolerance.y) position_constraint.constraint_region_shape.dimensions.append(2*pose_constraint.absolute_position_tolerance.z) position_constraint.constraint_region_orientation.x = 0.0 position_constraint.constraint_region_orientation.y = 0.0 position_constraint.constraint_region_orientation.z = 0.0 position_constraint.constraint_region_orientation.w = 1.0 position_constraint.weight = 1.0 orientation_constraint.header = pose_constraint.header orientation_constraint.link_name = pose_constraint.link_name orientation_constraint.orientation = pose_constraint.pose.orientation orientation_constraint.type = pose_constraint.orientation_constraint_type orientation_constraint.absolute_roll_tolerance = pose_constraint.absolute_roll_tolerance orientation_constraint.absolute_pitch_tolerance = pose_constraint.absolute_pitch_tolerance orientation_constraint.absolute_yaw_tolerance = pose_constraint.absolute_yaw_tolerance orientation_constraint.weight = 1.0 return position_constraint, orientation_constraint def add_goal_constraint_to_move_arm_goal(pose_constraint, move_arm_goal): position_constraint, orientation_constraint = pose_constraint_to_position_orientation_constraints(pose_constraint) move_arm_goal.motion_plan_request.goal_constraints.position_constraints.append(position_constraint) move_arm_goal.motion_plan_request.goal_constraints.orientation_constraints.append(orientation_constraint) if __name__ == '__main__': rospy.init_node('move_arm_pose_goal_test') move_arm = actionlib.SimpleActionClient('move_right_arm', MoveArmAction) move_arm.wait_for_server() rospy.loginfo('Connected to server') goalA = MoveArmGoal() goalA.motion_plan_request.group_name = 'right_arm' goalA.motion_plan_request.num_planning_attempts = 1 goalA.motion_plan_request.planner_id = '' goalA.planner_service_name = 'ompl_planning/plan_kinematic_path' goalA.motion_plan_request.allowed_planning_time = rospy.Duration(5.0) desired_pose = SimplePoseConstraint() desired_pose.header.frame_id = 'torso_lift_link' desired_pose.link_name = 'r_gripper_l_fingertip_link' desired_pose.pose.position.x = 0.75 desired_pose.pose.position.y = -0.188 desired_pose.pose.position.z = 0 desired_pose.pose.orientation.x = 0.0 desired_pose.pose.orientation.y = 0.0 desired_pose.pose.orientation.z = 0.0 desired_pose.pose.orientation.w = 1.0 desired_pose.absolute_position_tolerance.x = 0.02 desired_pose.absolute_position_tolerance.y = 0.02 desired_pose.absolute_position_tolerance.z = 0.02 desired_pose.absolute_roll_tolerance = 0.04 desired_pose.absolute_pitch_tolerance = 0.04 desired_pose.absolute_yaw_tolerance = 0.04 add_goal_constraint_to_move_arm_goal(desired_pose, goalA) move_arm.send_goal(goalA) finished_within_time = move_arm.wait_for_result(rospy.Duration(200.0)) if not finished_within_time: move_arm.cancel_goal() rospy.loginfo("Timed out achieving goal A") else: state = move_arm.get_state() if state == GoalStatus.SUCCEEDED: rospy.loginfo('Action finished and was successful.') else: rospy.loginfo('Action failed: %d'%(state))
[ [ 1, 0, 0.2353, 0.0074, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.2353, 0.0074, 0, 0.66, 0.0769, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.25, 0.0074, 0, 0.66...
[ "import roslib; roslib.load_manifest('move_arm_tutorials')", "import roslib; roslib.load_manifest('move_arm_tutorials')", "import rospy", "import actionlib", "import geometric_shapes_msgs", "from move_arm_msgs.msg import MoveArmAction", "from move_arm_msgs.msg import MoveArmGoal", "from motion_plannin...
# # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # # Author: Advait Jain (advait@cc.gatech.edu), Healthcare Robotics Lab, Georgia Tech import roslib; roslib.load_manifest('move_arm_tutorials') import rospy import actionlib import geometric_shapes_msgs from move_arm_msgs.msg import MoveArmAction from move_arm_msgs.msg import MoveArmGoal from motion_planning_msgs.msg import JointConstraint from actionlib_msgs.msg import GoalStatus if __name__ == '__main__': import hrl_lib.transforms as tr rospy.init_node('move_arm_joint_goal_test') move_arm = actionlib.SimpleActionClient('move_right_arm', MoveArmAction) move_arm.wait_for_server() rospy.loginfo('Connected to server') goalB = MoveArmGoal() names = ['r_shoulder_pan_joint', 'r_shoulder_lift_joint', 'r_upper_arm_roll_joint', 'r_elbow_flex_joint', 'r_forearm_roll_joint', 'r_wrist_flex_joint', 'r_wrist_roll_joint'] goalB.motion_plan_request.group_name = 'right_arm' goalB.motion_plan_request.num_planning_attempts = 1 goalB.motion_plan_request.allowed_planning_time = rospy.Duration(5.0) goalB.motion_plan_request.planner_id = '' goalB.planner_service_name = 'ompl_planning/plan_kinematic_path' import roslib; roslib.load_manifest('darpa_m3') import sandbox_advait.pr2_arms as pa pr2_arms = pa.PR2Arms() raw_input('Move arm to goal location and hit ENTER') q = pr2_arms.get_joint_angles(0) raw_input('Move arm to start location and hit ENTER') q[6] = tr.angle_within_mod180(q[6]) q[4] = tr.angle_within_mod180(q[4]) for i in range(7): jc = JointConstraint() jc.joint_name = names[i] jc.position = q[i] jc.tolerance_below = 0.1 jc.tolerance_above = 0.1 goalB.motion_plan_request.goal_constraints.joint_constraints.append(jc) move_arm.send_goal(goalB) finished_within_time = move_arm.wait_for_result(rospy.Duration(200.0)) if not finished_within_time: move_arm.cancel_goal() rospy.loginfo("Timed out achieving goal A") else: state = move_arm.get_state() if state == GoalStatus.SUCCEEDED: rospy.loginfo('Action finished and was successful.') else: rospy.loginfo('Action failed: %d'%(state))
[ [ 1, 0, 0.3137, 0.0098, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.3137, 0.0098, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.3333, 0.0098, 0, 0....
[ "import roslib; roslib.load_manifest('move_arm_tutorials')", "import roslib; roslib.load_manifest('move_arm_tutorials')", "import rospy", "import actionlib", "import geometric_shapes_msgs", "from move_arm_msgs.msg import MoveArmAction", "from move_arm_msgs.msg import MoveArmGoal", "from motion_plannin...
import roslib; roslib.load_manifest('hrl_pr2_kinematics_tutorials') import rospy import numpy as np, math ## Class defining the core EPC function and a few simple examples. # More complex behaviors that use EPC should have their own ROS # packages. class EPC(): def __init__(self, robot): self.robot = robot ## # @param equi_pt_generator: function that returns stop, ea where ea: equilibrium angles and stop: string which is '' for epc motion to continue # @param rapid_call_func: called in the time between calls to the equi_pt_generator can be used for logging, safety etc. returns string which is '' for epc motion to continue # @param time_step: time between successive calls to equi_pt_generator # @param arg_list - list of arguments to be passed to the equi_pt_generator # @return stop (the string which has the reason why the epc # motion stopped.), ea (last commanded equilibrium angles) def epc_motion(self, equi_pt_generator, time_step, arm, arg_list, rapid_call_func=None, control_function=None): stop, ea = equi_pt_generator(*arg_list) t_end = rospy.get_time() while stop == '': t_end += time_step #self.robot.set_jointangles(arm, ea) #import pdb; pdb.set_trace() control_function(arm, *ea) # self.robot.step() this should be within the rapid_call_func for the meka arms. t1 = rospy.get_time() while t1<t_end: if rapid_call_func != None: stop = rapid_call_func(arm) if stop != '': break # self.robot.step() this should be within the rapid_call_func for the meka arms. t1 = rospy.get_time() stop, ea = equi_pt_generator(*arg_list) if stop == 'reset timing': stop = '' t_end = rospy.get_time() return stop, ea ## Pull back along a straight line (-ve x direction) # @param arm - 'right_arm' or 'left_arm' # @param ea - starting equilibrium angle. # @param rot_mat - rotation matrix defining end effector pose # @param distance - how far back to pull. def pull_back(self, arm, ea, rot_mat, distance): self.cep = self.robot.FK(arm, ea) self.dist_left = distance self.ea = ea def eq_gen_pull_back(robot, arm, rot_mat): if self.dist_left <= 0.: return 'done', None step_size = 0.01 self.cep[0,0] -= step_size self.dist_left -= step_size ea = robot.IK(arm, self.cep, rot_mat, self.ea) self.ea = ea if ea == None: return 'IK fail', ea return '', [ea,] arg_list = [self.robot, arm, rot_mat] stop, ea = self.epc_motion(eq_gen_pull_back, 0.1, arm, arg_list, control_function = self.robot.set_jointangles) print stop, ea ## Pull back along a straight line (-ve x direction) # @param arm - 'right_arm' or 'left_arm' # @param ea - starting cep. # @param rot_mat - rotation matrix defining end effector pose # @param distance - how far back to pull. def pull_back_cartesian_control(self, arm, cep, rot_mat, distance): self.cep = cep self.dist_left = distance def eq_gen_pull_back(robot, arm, rot_mat): if self.dist_left <= 0.: return 'done', None step_size = 0.01 self.cep[0,0] -= step_size self.dist_left -= step_size if self.cep[0,0] < 0.4: return 'very close to the body: %.3f'%self.cep[0,0], None return '', (self.cep, rot_mat) arg_list = [self.robot, arm, rot_mat] stop, ea = self.epc_motion(eq_gen_pull_back, 0.1, arm, arg_list, control_function = self.robot.set_cartesian) print stop, ea if __name__ == '__main__': import hrl_pr2 import hrl_lib.transforms as tr rospy.init_node('epc_pr2', anonymous = True) rospy.logout('epc_pr2: ready') pr2 = hrl_pr2.HRL_PR2() epc = EPC(pr2) arm = 'right_arm' if False: ea = [0, 0, 0, 0, 0, 0, 0] ea = epc.robot.get_joint_angles(arm) rospy.logout('Going to starting position') epc.robot.set_jointangles(arm, ea, duration=4.0) raw_input('Hit ENTER to pull') epc.pull_back(arm, ea, tr.Rx(0), 0.2) if True: p = np.matrix([0.9, -0.3, -0.15]).T rot = tr.Rx(0.) rot = tr.Rx(math.radians(90.)) rospy.logout('Going to starting position') # epc.robot.open_gripper(arm) epc.robot.set_cartesian(arm, p, rot) # raw_input('Hit ENTER to close the gripper') # epc.robot.close_gripper(arm) raw_input('Hit ENTER to pull') epc.pull_back_cartesian_control(arm, p, rot, 0.4)
[ [ 1, 0, 0.0145, 0.0072, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0145, 0.0072, 0, 0.66, 0.2, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0217, 0.0072, 0, 0.66,...
[ "import roslib; roslib.load_manifest('hrl_pr2_kinematics_tutorials')", "import roslib; roslib.load_manifest('hrl_pr2_kinematics_tutorials')", "import rospy", "import numpy as np, math", "class EPC():\n def __init__(self, robot):\n self.robot = robot\n\n ##\n # @param equi_pt_generator: funct...
import numpy as np, math from threading import RLock import roslib; roslib.load_manifest('hrl_pr2_kinematics_tutorials') import rospy import actionlib from kinematics_msgs.srv import GetPositionFK, GetPositionFKRequest, GetPositionFKResponse from kinematics_msgs.srv import GetPositionIK, GetPositionIKRequest, GetPositionIKResponse from pr2_controllers_msgs.msg import JointTrajectoryAction, JointTrajectoryGoal, JointTrajectoryControllerState from pr2_controllers_msgs.msg import Pr2GripperCommandGoal, Pr2GripperCommandAction, Pr2GripperCommand from trajectory_msgs.msg import JointTrajectoryPoint from geometry_msgs.msg import PoseStamped from std_msgs.msg import Float64 from sensor_msgs.msg import JointState import hrl_lib.transforms as tr import time class HRL_PR2(): def __init__(self): self.joint_names_list = ['r_shoulder_pan_joint', 'r_shoulder_lift_joint', 'r_upper_arm_roll_joint', 'r_elbow_flex_joint', 'r_forearm_roll_joint', 'r_wrist_flex_joint', 'r_wrist_roll_joint'] rospy.wait_for_service('pr2_right_arm_kinematics/get_fk'); rospy.wait_for_service('pr2_right_arm_kinematics/get_ik'); self.fk_srv = rospy.ServiceProxy('pr2_right_arm_kinematics/get_fk', GetPositionFK) self.ik_srv = rospy.ServiceProxy('pr2_right_arm_kinematics/get_ik', GetPositionIK) self.joint_action_client = actionlib.SimpleActionClient('r_arm_controller/joint_trajectory_action', JointTrajectoryAction) self.gripper_action_client = actionlib.SimpleActionClient('r_gripper_controller/gripper_action', Pr2GripperCommandAction) self.joint_action_client.wait_for_server() self.gripper_action_client.wait_for_server() self.arm_state_lock = RLock() #rospy.Subscriber('/r_arm_controller/state', JointTrajectoryControllerState, self.r_arm_state_cb) rospy.Subscriber('/joint_states', JointState, self.joint_states_cb) self.r_arm_cart_pub = rospy.Publisher('/r_cart/command_pose', PoseStamped) self.r_arm_pub_l = [] self.joint_nm_list = ['shoulder_pan', 'shoulder_lift', 'upper_arm_roll', 'elbow_flex', 'forearm_roll', 'wrist_flex', 'wrist_roll'] self.r_arm_angles = None self.r_arm_efforts = None for nm in self.joint_nm_list: self.r_arm_pub_l.append(rospy.Publisher('r_'+nm+'_controller/command', Float64)) rospy.sleep(1.) def joint_states_cb(self, data): r_arm_angles = [] r_arm_efforts = [] r_jt_idx_list = [17, 18, 16, 20, 19, 21, 22] for i,nm in enumerate(self.joint_nm_list): idx = r_jt_idx_list[i] if data.name[idx] != 'r_'+nm+'_joint': raise RuntimeError('joint angle name does not match. Expected: %s, Actual: %s i: %d'%('r_'+nm+'_joint', data.name[idx], i)) r_arm_angles.append(data.position[idx]) r_arm_efforts.append(data.effort[idx]) self.arm_state_lock.acquire() self.r_arm_angles = r_arm_angles self.r_arm_efforts = r_arm_efforts self.arm_state_lock.release() ## go to a joint configuration. # @param q - list of 7 joint angles in RADIANS. # @param duration - how long (SECONDS) before reaching the joint angles. def set_jointangles(self, arm, q, duration=0.15): rospy.logwarn('Currently ignoring the arm parameter.') # for i,p in enumerate(self.r_arm_pub_l): # p.publish(q[i]) jtg = JointTrajectoryGoal() jtg.trajectory.joint_names = self.joint_names_list jtp = JointTrajectoryPoint() jtp.positions = q jtp.velocities = [0 for i in range(len(q))] jtp.accelerations = [0 for i in range(len(q))] jtp.time_from_start = rospy.Duration(duration) jtg.trajectory.points.append(jtp) self.joint_action_client.send_goal(jtg) def FK(self, arm, q): rospy.logwarn('Currently ignoring the arm parameter.') fk_req = GetPositionFKRequest() fk_req.header.frame_id = 'torso_lift_link' fk_req.fk_link_names.append('r_wrist_roll_link') fk_req.robot_state.joint_state.name = self.joint_names_list fk_req.robot_state.joint_state.position = q fk_resp = GetPositionFKResponse() fk_resp = self.fk_srv.call(fk_req) if fk_resp.error_code.val == fk_resp.error_code.SUCCESS: x = fk_resp.pose_stamped[0].pose.position.x y = fk_resp.pose_stamped[0].pose.position.y z = fk_resp.pose_stamped[0].pose.position.z ret = np.matrix([x,y,z]).T else: rospy.logerr('Forward kinematics failed') ret = None return ret def IK(self, arm, p, rot, q_guess): rospy.logwarn('Currently ignoring the arm parameter.') ik_req = GetPositionIKRequest() ik_req.timeout = rospy.Duration(5.) ik_req.ik_request.ik_link_name = 'r_wrist_roll_link' ik_req.ik_request.pose_stamped.header.frame_id = 'torso_lift_link' ik_req.ik_request.pose_stamped.pose.position.x = p[0,0] ik_req.ik_request.pose_stamped.pose.position.y = p[1,0] ik_req.ik_request.pose_stamped.pose.position.z = p[2,0] quat = tr.matrix_to_quaternion(rot) ik_req.ik_request.pose_stamped.pose.orientation.x = quat[0] ik_req.ik_request.pose_stamped.pose.orientation.y = quat[1] ik_req.ik_request.pose_stamped.pose.orientation.z = quat[2] ik_req.ik_request.pose_stamped.pose.orientation.w = quat[3] ik_req.ik_request.ik_seed_state.joint_state.position = q_guess ik_req.ik_request.ik_seed_state.joint_state.name = self.joint_names_list ik_resp = self.ik_srv.call(ik_req) if ik_resp.error_code.val == ik_resp.error_code.SUCCESS: ret = ik_resp.solution.joint_state.position else: rospy.logerr('Inverse kinematics failed') ret = None return ret # for compatibility with Meka arms on Cody. Need to figure out a # good way to have a common interface to different arms. def step(self): return def end_effector_pos(self, arm): q = self.get_joint_angles(arm) return self.FK(arm, q) def get_joint_angles(self, arm): rospy.logwarn('Currently ignoring the arm parameter.') self.arm_state_lock.acquire() q = self.r_arm_angles self.arm_state_lock.release() return q # need for search and hook def go_cartesian(self, arm): rospy.logerr('Need to implement this function.') raise RuntimeError('Unimplemented function') def get_wrist_force(self, arm, bias = True, base_frame = False): rospy.logerr('Need to implement this function.') raise RuntimeError('Unimplemented function') def set_cartesian(self, arm, p, rot): rospy.logwarn('Currently ignoring the arm parameter.') ps = PoseStamped() ps.header.stamp = rospy.rostime.get_rostime() ps.header.frame_id = 'torso_lift_link' ps.pose.position.x = p[0,0] ps.pose.position.y = p[1,0] ps.pose.position.z = p[2,0] quat = tr.matrix_to_quaternion(rot) ps.pose.orientation.x = quat[0] ps.pose.orientation.y = quat[1] ps.pose.orientation.z = quat[2] ps.pose.orientation.w = quat[3] self.r_arm_cart_pub.publish(ps) def open_gripper(self, arm): self.gripper_action_client.send_goal(Pr2GripperCommandGoal(Pr2GripperCommand(position=0.08, max_effort = -1))) ## close the gripper # @param effort - supposed to be in Newtons. (-ve number => max effort) def close_gripper(self, arm, effort = 15): self.gripper_action_client.send_goal(Pr2GripperCommandGoal(Pr2GripperCommand(position=0.0, max_effort = effort))) def get_wrist_force(self, arm): pass if __name__ == '__main__': rospy.init_node('hrl_pr2', anonymous = True) rospy.logout('hrl_pr2: ready') hrl_pr2 = HRL_PR2() if False: q = [0, 0, 0, 0, 0, 0, 0] hrl_pr2.set_jointangles('right_arm', q) ee_pos = hrl_pr2.FK('right_arm', q) print 'FK result:', ee_pos.A1 ee_pos[0,0] -= 0.1 q_ik = hrl_pr2.IK('right_arm', ee_pos, tr.Rx(0.), q) print 'q_ik:', [math.degrees(a) for a in q_ik] rospy.spin() if False: p = np.matrix([0.9, -0.3, -0.15]).T #rot = tr.Rx(0.) rot = tr.Rx(math.radians(90.)) hrl_pr2.set_cartesian('right_arm', p, rot) hrl_pr2.open_gripper('right_arm') raw_input('Hit ENTER to close') hrl_pr2.close_gripper('right_arm', effort = 15)
[ [ 1, 0, 0.0089, 0.0044, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0133, 0.0044, 0, 0.66, 0.0588, 83, 0, 1, 0, 0, 83, 0, 0 ], [ 1, 0, 0.0222, 0.0044, 0, 0....
[ "import numpy as np, math", "from threading import RLock", "import roslib; roslib.load_manifest('hrl_pr2_kinematics_tutorials')", "import roslib; roslib.load_manifest('hrl_pr2_kinematics_tutorials')", "import rospy", "import actionlib", "from kinematics_msgs.srv import GetPositionFK, GetPositionFKReques...
import roslib; roslib.load_manifest('pr2_doors_epc') import rospy import hrl_lib.transforms as tr import force_torque.FTClient as ftc import math, numpy as np from rviz_marker_test import * from hrl_msgs.msg import FloatArray if __name__ == '__main__': ft_client = ftc.FTClient('force_torque_ft2') ft_client.bias() marker_pub = rospy.Publisher('/test_marker', Marker) ati_pub = rospy.Publisher('/ati_ft', FloatArray) p_st = np.matrix([0.,0.,0.]).T force_frame_id = 'r_wrist_roll_link' while not rospy.is_shutdown(): ft = ft_client.read(fresh = True) rmat = tr.Rx(math.radians(180.)) * tr.Ry(math.radians(-90.)) * tr.Rz(math.radians(30.)) force = rmat * ft[0:3,:] print 'Force:', force.A1 # force is now in the 'robot' coordinate frame. force_scale = 0.1 p_end = p_st + force * force_scale marker = get_marker_arrow(p_st, p_end, force_frame_id) marker_pub.publish(marker) farr = FloatArray() farr.header.stamp = rospy.rostime.get_rostime() farr.header.frame_id = force_frame_id farr.data = (-force).A1.tolist() ati_pub.publish(farr) # rospy.sleep(0.1)
[ [ 1, 0, 0.0526, 0.0263, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0526, 0.0263, 0, 0.66, 0.125, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0789, 0.0263, 0, 0.6...
[ "import roslib; roslib.load_manifest('pr2_doors_epc')", "import roslib; roslib.load_manifest('pr2_doors_epc')", "import rospy", "import hrl_lib.transforms as tr", "import force_torque.FTClient as ftc", "import math, numpy as np", "from rviz_marker_test import *", "from hrl_msgs.msg import FloatArray",...
import roslib; roslib.load_manifest('hrl_pr2_kinematics') import rospy from equilibrium_point_control.msg import MechanismKinematicsRot from equilibrium_point_control.msg import MechanismKinematicsJac import epc from threading import RLock ## # compute the end effector rotation matrix. # @param hook - hook angle. RADIANS(0, -90, 90) (hor, up, down) # @param angle - angle between robot and surface normal. # Angle about the Z axis through which the robot must turn to face # the surface. def rot_mat_from_angles(hook, surface): rot_mat = tr.Rz(hook) * tr.Rx(surface) return rot_mat class Door_EPC(epc.EPC): def __init__(self, robot): epc.EPC.__init__(self, robot) self.mech_kinematics_lock = RLock() self.fit_circle_lock = RLock() rospy.Subscriber('mechanism_kinematics_rot', MechanismKinematicsRot, self.mechanism_kinematics_rot_cb) rospy.Subscriber('mechanism_kinematics_jac', MechanismKinematicsJac, self.mechanism_kinematics_jac_cb) self.mech_traj_pub = rospy.Publisher('mechanism_trajectory', Point32) self.eq_pt_not_moving_counter = 0 ## log the joint angles, equi pt joint angles and forces. def log_state(self, arm): t_now = rospy.get_time() q_now = self.robot.get_joint_angles(arm) self.pull_trajectory.q_list.append(q_now) self.pull_trajectory.time_list.append(t_now) self.eq_pt_trajectory.q_list.append(self.q_guess) # see equi_pt_generator - q_guess is the config for the eq point. self.eq_pt_trajectory.time_list.append(t_now) wrist_force = self.robot.get_wrist_force(arm, base_frame=True) self.force_trajectory.f_list.append(wrist_force.A1.tolist()) self.force_trajectory.time_list.append(t_now) return '' # log_state also used as a rapid_call_func ## # @param arm - 'right_arm' or 'left_arm' # @param motion vec is in tl frame. # @param step_size - distance (meters) through which CEP should move # @param rot_mat - rotation matrix for IK # @return JEP def update_eq_point(self, arm, motion_vec, step_size, rot_mat): self.eq_pt_cartesian = self.eq_pt_cartesian_ts next_pt = self.eq_pt_cartesian + step_size * motion_vec q_eq = self.robot.IK(arm, next_pt, rot_mat, self.q_guess) self.eq_pt_cartesian = next_pt self.eq_pt_cartesian_ts = self.eq_pt_cartesian self.q_guess = q_eq return q_eq def common_stopping_conditions(self): stop = '' if self.q_guess == None: stop = 'IK fail' wrist_force = self.robot.get_wrist_force('right_arm',base_frame=True) mag = np.linalg.norm(wrist_force) print 'force magnitude:', mag if mag > self.eq_force_threshold: stop = 'force exceed' if mag < 1.2 and self.hooked_location_moved: if (self.prev_force_mag - mag) > 30.: stop = 'slip: force step decrease and below thresold.' #stop = '' else: self.slip_count += 1 else: self.slip_count = 0 if self.slip_count == 10: stop = 'slip: force below threshold for too long.' return stop ## constantly update the estimate of the kinematics and move the # equilibrium point along the tangent of the estimated arc, and # try to keep the radial force constant. # @param h_force_possible - True (hook side) or False (hook up). # @param v_force_possible - False (hook side) or True (hook up). # Is maintaining a radial force possible or not (based on hook # geometry and orientation) # @param cep_vel - tangential velocity of the cep in m/s def eqpt_gen_control_radial_force(self, arm, rot_mat, h_force_possible, v_force_possible, cep_vel): self.log_state(arm) step_size = 0.1 * cep_vel # 0.1 is the time interval between calls to the equi_generator function (see pull) q_eq = self.update_eq_point(arm, self.eq_motion_vec, step_size, rot_mat) stop = self.common_stopping_conditions() wrist_force = self.robot.get_wrist_force(arm, base_frame=True) mag = np.linalg.norm(wrist_force) curr_pos = self.robot.FK(arm,self.pull_trajectory.q_list[-1]) if len(self.pull_trajectory.q_list)>1: start_pos = np.matrix(self.cartesian_pts_list[0]).T else: start_pos = curr_pos #mechanism kinematics. self.mech_traj_pub.publish(Point32(curr_pos[0,0], curr_pos[1,0], curr_pos[2,0])) if self.use_jacobian: self.mech_kinematics_lock.acquire() self.cartesian_pts_list.append(curr_pos.A1.tolist()) tangential_vec_ts = self.tangential_vec_ts force_vec_ts = self.constraint_vec_1_ts + self.constraint_vec_2_ts # this is specifically for the right arm, and lots of # other assumptions about the hook etc. (Advait, Feb 28, 2010) if h_force_possible: force_vec_ts[2,0] = 0. if v_force_possible: force_vec_ts[1,0] = 0. force_vec_ts = force_vec_ts / np.linalg.norm(force_vec_ts) if force_vec_ts[2,0] < 0.: # only allowing upward force. force_vec_ts = -force_vec_ts if force_vec_ts[1,0] < 0.: # only allowing force to the left force_vec_ts = -force_vec_ts self.mech_kinematics_lock.release() force_vec = force_vec_ts tangential_vec = tangential_vec_ts if self.use_rotation_center: self.fit_circle_lock.acquire() self.cartesian_pts_list.append(curr_pos.A1.tolist()) rad = self.rad cx_start, cy_start = self.cx_start, self.cy_start cz_start = self.cz_start self.fit_circle_lock.release() cx, cy = cx_start, cy_start cz = cz_start radial_vec = curr_pos_tl-np.matrix([cx,cy,cz]).T radial_vec = radial_vec/np.linalg.norm(radial_vec) if cy_start<start_pos[1,0]: tan_x,tan_y = -radial_vec[1,0],radial_vec[0,0] else: tan_x,tan_y = radial_vec[1,0],-radial_vec[0,0] if tan_x > 0. and (start_pos[0,0]-curr_pos[0,0]) < 0.09: tan_x = -tan_x tan_y = -tan_y if cy_start > start_pos[1,0]: radial_vec = -radial_vec # axis to the left, want force in # anti-radial direction. rv = radial_vec force_vec = np.matrix([rv[0,0], rv[1,0], 0.]).T tangential_vec = np.matrix([tan_x, tan_y, 0.]).T tangential_vec_ts = tangential_vec radial_vec_ts = radial_vec force_vec_ts = force_vec f_vec = -1*np.array([wrist_force[0,0], wrist_force[1,0], wrist_force[2,0]]) f_rad_mag = np.dot(f_vec, force_vec.A1) err = f_rad_mag-5. if err>0.: kp = -0.1 else: kp = -0.2 radial_motion_mag = kp * err # radial_motion_mag in cm (depends on eq_motion step size) if self.use_rotation_center: if h_force_possible == False and parallel_to_floor: radial_motion_mag = 0. if v_force_possible == False and parallel_to_floor == False: radial_motion_mag = 0. radial_motion_vec = force_vec * radial_motion_mag self.eq_motion_vec = copy.copy(tangential_vec) self.eq_motion_vec += radial_motion_vec self.prev_force_mag = mag if self.init_tangent_vector == None: self.init_tangent_vector = copy.copy(tangential_vec_ts) c = np.dot(tangential_vec_ts.A1, self.init_tangent_vector.A1) ang = np.arccos(c) dist_moved = np.dot((curr_pos - start_pos).A1, self.tangential_vec_ts.A1) ftan = abs(np.dot(wrist_force.A1, tangential_vec.A1)) if abs(dist_moved) > 0.09 and self.hooked_location_moved == False: # change the force threshold once the hook has started pulling. self.hooked_location_moved = True self.eq_force_threshold = ut.bound(mag+30.,20.,80.) self.ftan_threshold = self.ftan_threshold + max(10., 1.5*ftan) if self.hooked_location_moved: if abs(tangential_vec_ts[2,0]) < 0.2 and ftan > self.ftan_threshold: stop = 'ftan threshold exceed: %f'%ftan else: self.ftan_threshold = max(self.ftan_threshold, ftan) if self.hooked_location_moved and ang > math.radians(90.): print 'Angle:', math.degrees(ang) self.open_ang_exceed_count += 1 if self.open_ang_exceed_count > 2: stop = 'opened mechanism through large angle: %.1f'%(math.degrees(ang)) else: self.open_ang_exceed_count = 0 self.mech_time_list.append(time.time()) self.mech_x_list.append(ang) self.f_rad_list.append(f_rad_mag) self.f_tan_list.append(np.dot(f_vec, tangential_vec.A1)) self.tan_vec_list.append(tangential_vec_ts.A1.tolist()) self.rad_vec_list.append(force_vec_ts.A1.tolist()) return stop, q_eq def pull(self, arm, hook_angle, force_threshold, ea, kinematics_estimation = 'rotation_center', pull_left = False): self.init_tangent_vector = None self.open_ang_exceed_count = 0. if kinematics_estimation == 'rotation_center': self.use_rotation_center = True else: self.use_rotation_center = False if kinematics_estimation == 'jacobian': self.use_jacobian = True else: self.use_jacobian = False #rot_mat = tr.Rz(hook_angle)*tr.Ry(math.radians(-90)) rot_mat = rot_mat_from_angles(hook_angle, surface_angle) time_dict = {} time_dict['before_pull'] = time.time() self.pull_trajectory = at.JointTrajectory() self.eq_pt_trajectory = at.JointTrajectory() self.force_trajectory = at.ForceTrajectory() self.robot.step() start_config = self.robot.get_joint_angles(arm) self.eq_force_threshold = force_threshold self.ftan_threshold = 2. self.hooked_location_moved = False # flag to indicate when the hooking location started moving. self.prev_force_mag = np.linalg.norm(self.robot.get_wrist_force(arm)) self.eq_motion_vec = np.matrix([-1.,0.,0.]).T # might want to change this to account for the surface_angle. self.slip_count = 0 self.eq_pt_cartesian = self.robot.FK(arm, ea) self.eq_pt_cartesian_ts = self.robot.FK(arm, ea) self.start_pos = copy.copy(self.eq_pt_cartesian) self.q_guess = ea if not pull_left: self.tangential_vec_ts = np.matrix([-1.,0.,0.]).T self.constraint_vec_2_ts = np.matrix([0.,0.,1.]).T self.constraint_vec_1_ts = np.matrix([0.,1.,0.]).T else: self.tangential_vec_ts = np.matrix([0.,1.,0.]).T self.constraint_vec_2_ts = np.matrix([0.,0.,1.]).T self.constraint_vec_1_ts = np.matrix([1.,0.,0.]).T self.mech_time_list = [] self.mech_x_list = [] self.f_rad_list = [] self.f_tan_list = [] self.tan_vec_list = [] self.rad_vec_list = [] self.cartesian_pts_list = [] ee_pos = self.robot.end_effector_pos(arm) if self.use_rotation_center: # this might have to change depending on left and right # arm? or maybe not since the right arm can open both # doors. self.cx_start = ee_pos[0,0] self.cy_start = ee_pos[1,0]-1.0 self.cz_start = ee_pos[2,0] self.rad = 5.0 h_force_possible = abs(hook_angle) < math.radians(30.) v_force_possible = abs(hook_angle) > math.radians(60.) arg_list = [arm, rot_mat, h_force_possible, v_force_possible, cep_vel] result, jep = self.epc_motion(self.eqpt_gen_control_radial_force, 0.1, arm, arg_list, self.log_state) print 'EPC motion result:', result print 'Original force threshold:', force_threshold print 'Adapted force threshold:', self.eq_force_threshold print 'Adapted ftan threshold:', self.ftan_threshold time_dict['after_pull'] = time.time() d = {'actual': self.pull_trajectory, 'eq_pt': self.eq_pt_trajectory, 'force': self.force_trajectory, 'torque': self.jt_torque_trajectory, 'info': info_string, 'force_threshold': force_threshold, 'eq_force_threshold': self.eq_force_threshold, 'hook_angle':hook_angle, 'result':result, 'time_dict':time_dict, 'cep_vel': cep_vel, 'ftan_threshold': self.ftan_threshold} self.robot.step() ut.save_pickle(d,'pull_trajectories_'+d['info']+'_'+ut.formatted_time()+'.pkl') dd = {'mechanism_x': self.mech_x_list, 'mechanism_time': self.mech_time_list, 'force_rad_list': self.f_rad_list, 'force_tan_list': self.f_tan_list, 'tan_vec_list': self.tan_vec_list, 'rad_vec_list': self.rad_vec_list } ut.save_pickle(dd,'mechanism_trajectories_robot_'+d['info']+'_'+ut.formatted_time()+'.pkl') ## behavior to search around the hook_loc to try and get a good # hooking grasp # @param arm - 'right_arm' or 'left_arm' # @param hook_angle - radians(0,-90,90) side,up,down # @param hook_loc - 3x1 np matrix # @param angle - angle between torso x axis and surface normal. # @return s, jep (stopping string and last commanded JEP) def search_and_hook(self, arm, hook_angle, hook_loc, angle, hooking_force_threshold = 5.): rot_mat = rot_mat_from_angles(hook_angle, angle) if arm == 'right_arm': hook_dir = np.matrix([0.,1.,0.]).T # hook direc in home position elif arm == 'left_arm': hook_dir = np.matrix([0.,-1.,0.]).T # hook direc in home position else: raise RuntimeError('Unknown arm: %s', arm) start_loc = hook_loc + rot_mat.T * hook_dir * -0.03 # 3cm direc opposite to hook. # vector normal to surface and pointing into the surface. normal_tl = tr.Rz(-angle) * np.matrix([1.0,0.,0.]).T pt1 = start_loc - normal_tl * 0.1 pt1[2,0] -= 0.02 # funny error in meka control code? or is it gravity comp? self.robot.go_cartesian(arm, pt1, rot_mat, speed=0.2) vec = normal_tl * 0.2 s, jep = self.firenze.move_till_hit(arm, vec=vec, force_threshold=2.0, rot=rot_mat, speed=0.07) self.eq_pt_cartesian = self.firenze.FK(arm, jep) self.eq_pt_cartesian_ts = self.firenze.FK(arm, jep) self.start_pos = copy.copy(self.eq_pt_cartesian) self.q_guess = jep move_dir = rot_mat.T * hook_dir arg_list = [arm, move_dir, rot_mat, hooking_force_threshold] s, jep = self.compliant_motion(self.equi_generator_surface_follow, 0.05, arm, arg_list) return s, jep if __name__ == '__main__': print 'Hello World'
[ [ 1, 0, 0.0052, 0.0026, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0052, 0.0026, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0078, 0.0026, 0, 0....
[ "import roslib; roslib.load_manifest('hrl_pr2_kinematics')", "import roslib; roslib.load_manifest('hrl_pr2_kinematics')", "import rospy", "from equilibrium_point_control.msg import MechanismKinematicsRot", "from equilibrium_point_control.msg import MechanismKinematicsJac", "import epc", "from threading ...
__all__ = [ 'door_epc', 'epc', 'hrl_pr2', ]
[ [ 14, 0, 0.625, 0.625, 0, 0.66, 0, 272, 0, 0, 0, 0, 0, 5, 0 ] ]
[ "__all__ = [\n'door_epc',\n'epc',\n'hrl_pr2',\n]" ]
import roslib; roslib.load_manifest('hrl_pr2_kinematics_tutorials') import rospy from visualization_msgs.msg import Marker from geometry_msgs.msg import Point import hrl_lib.transforms as tr import math, numpy as np def get_marker_arrow(p_st, p_end, frame_id): m = Marker() m.header.stamp = rospy.rostime.get_rostime() m.header.frame_id = frame_id m.ns = 'basic_shapes' m.id = 0 m.type = Marker.ARROW m.action = Marker.ADD pt1 = Point(p_st[0,0], p_st[1,0], p_st[2,0]) pt2 = Point(p_end[0,0], p_end[1,0], p_end[2,0]) m.points.append(pt1) m.points.append(pt2) m.scale.x = 0.02; m.scale.y = 0.05; m.scale.z = 0.1; m.color.r = 1.0; m.color.g = 0.0; m.color.b = 0.0; m.color.a = 1.0; m.lifetime = rospy.Duration(); return m if __name__ == '__main__': rospy.init_node('marker_test', anonymous = True) marker_pub = rospy.Publisher('/test_marker', Marker) p1 = np.matrix([0.,0.,0.]).T p2 = np.matrix([0.,1.,0.]).T while not rospy.is_shutdown(): marker = get_marker_arrow(p1, p2, 'base_link') marker_pub.publish(marker) rospy.sleep(0.1)
[ [ 1, 0, 0.0408, 0.0204, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0408, 0.0204, 0, 0.66, 0.125, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0612, 0.0204, 0, 0.6...
[ "import roslib; roslib.load_manifest('hrl_pr2_kinematics_tutorials')", "import roslib; roslib.load_manifest('hrl_pr2_kinematics_tutorials')", "import rospy", "from visualization_msgs.msg import Marker", "from geometry_msgs.msg import Point", "import hrl_lib.transforms as tr", "import math, numpy as np",...
# # # Copyright (c) 2010, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) import roslib roslib.load_manifest('hrl_simple_arm_goals') import rospy import actionlib from move_arm_msgs.msg import MoveArmGoal from move_arm_msgs.msg import MoveArmAction from motion_planning_msgs.msg import PositionConstraint from motion_planning_msgs.msg import OrientationConstraint from geometric_shapes_msgs.msg import Shape from actionlib_msgs.msg import GoalStatus if __name__ == '__main__': rospy.init_node('arm_cartesian_goal_sender') move_arm = actionlib.SimpleActionClient('move_right_arm', MoveArmAction) move_arm.wait_for_server() rospy.logout('Connected to server') goalA = MoveArmGoal() goalA.motion_plan_request.group_name = 'right_arm' goalA.motion_plan_request.num_planning_attempts = 1 goalA.motion_plan_request.planner_id = '' goalA.planner_service_name = 'ompl_planning/plan_kinematic_path' goalA.motion_plan_request.allowed_planning_time = rospy.Duration(5.) pc = PositionConstraint() pc.header.stamp = rospy.Time.now() pc.header.frame_id = 'torso_lift_link' pc.link_name = 'r_wrist_roll_link' pc.position.x = 0.75 pc.position.y = -0.188 pc.position.z = 0 pc.constraint_region_shape.type = Shape.BOX pc.constraint_region_shape.dimensions = [0.02, 0.02, 0.02] pc.constraint_region_orientation.w = 1.0 goalA.motion_plan_request.goal_constraints.position_constraints.append(pc) oc = OrientationConstraint() oc.header.stamp = rospy.Time.now() oc.header.frame_id = 'torso_lift_link' oc.link_name = 'r_wrist_roll_link' oc.orientation.x = 0. oc.orientation.y = 0. oc.orientation.z = 0. oc.orientation.w = 1. oc.absolute_roll_tolerance = 0.04 oc.absolute_pitch_tolerance = 0.04 oc.absolute_yaw_tolerance = 0.04 oc.weight = 1. goalA.motion_plan_request.goal_constraints.orientation_constraints.append(oc) move_arm.send_goal(goalA) finished_within_time = move_arm.wait_for_result() if not finished_within_time: move_arm.cancel_goal() rospy.logout('Timed out achieving goal A') else: state = move_arm.get_state() if state == GoalStatus.SUCCEEDED: rospy.logout('Action finished with SUCCESS') else: rospy.logout('Action failed')
[ [ 1, 0, 0.3048, 0.0095, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.3143, 0.0095, 0, 0.66, 0.1, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.3333, 0.0095, 0, 0.66,...
[ "import roslib", "roslib.load_manifest('hrl_simple_arm_goals')", "import rospy", "import actionlib", "from move_arm_msgs.msg import MoveArmGoal", "from move_arm_msgs.msg import MoveArmAction", "from motion_planning_msgs.msg import PositionConstraint", "from motion_planning_msgs.msg import OrientationC...
import matplotlib.pyplot as pp import numpy as np import roslib; roslib.load_manifest('hrl_pr2_door_opening') roslib.load_manifest('equilibrium_point_control') import hrl_lib.util as ut import equilibrium_point_control.arm_trajectories_ram as atr d = ut.load_pickle('pkls/ikea_cabinet_log.pkl') #d = ut.load_pickle('pkls/ikea_cabinet_2.pkl') #d = ut.load_pickle('pkls/lab_cabinet_log.pkl') typ = 'rotary' pr2_log = True d['f_list'] = d['f_list_estimate'] h_config, h_ftan_estimate = atr.force_trajectory_in_hindsight(d, typ, pr2_log) pp.plot(np.degrees(h_config), h_ftan_estimate, 'ro-', mew=0, ms=0, label='estimate') if 'f_list_torques' in d: d['f_list'] = d['f_list_torques'] h_config, h_ftan_torques = atr.force_trajectory_in_hindsight(d, typ, pr2_log) pp.plot(np.degrees(h_config), h_ftan_torques, 'go-', mew=0, ms=0, label='torques') d['f_list'] = d['f_list_ati'] h_config, h_ftan_ati = atr.force_trajectory_in_hindsight(d, typ, pr2_log) pp.plot(np.degrees(h_config), h_ftan_ati, 'bo-', mew=0, ms=0, label='ATI') pp.legend() pp.show()
[ [ 1, 0, 0.0513, 0.0256, 0, 0.66, 0, 596, 0, 1, 0, 0, 596, 0, 0 ], [ 1, 0, 0.0769, 0.0256, 0, 0.66, 0.0556, 954, 0, 1, 0, 0, 954, 0, 0 ], [ 1, 0, 0.1282, 0.0256, 0, ...
[ "import matplotlib.pyplot as pp", "import numpy as np", "import roslib; roslib.load_manifest('hrl_pr2_door_opening')", "import roslib; roslib.load_manifest('hrl_pr2_door_opening')", "roslib.load_manifest('equilibrium_point_control')", "import hrl_lib.util as ut", "import equilibrium_point_control.arm_tr...
import numpy as np, math from threading import RLock, Timer import sys, copy import roslib; roslib.load_manifest('hrl_pr2_lib') roslib.load_manifest('force_torque') # hack by Advait import force_torque.FTClient as ftc import tf import hrl_lib.transforms as tr import hrl_lib.viz as hv import rospy import PyKDL as kdl import actionlib from actionlib_msgs.msg import GoalStatus from kinematics_msgs.srv import GetPositionIK, GetPositionIKRequest, GetPositionIKResponse from pr2_controllers_msgs.msg import JointTrajectoryAction, JointTrajectoryGoal, JointTrajectoryControllerState from pr2_controllers_msgs.msg import Pr2GripperCommandGoal, Pr2GripperCommandAction, Pr2GripperCommand from trajectory_msgs.msg import JointTrajectoryPoint from geometry_msgs.msg import PoseStamped from teleop_controllers.msg import JTTeleopControllerState from std_msgs.msg import Float64 from sensor_msgs.msg import JointState import hrl_lib.transforms as tr import hrl_lib.kdl_utils as ku import time from visualization_msgs.msg import Marker node_name = "pr2_arms" def log(str): rospy.loginfo(node_name + ": " + str) ## # Convert arrays, lists, matricies to column format. # # @param x the unknown format # @return a column matrix def make_column(x): if (type(x) == type([]) or (type(x) == np.ndarray and x.ndim == 1) or type(x) == type(())): return np.matrix(x).T if type(x) == np.ndarray: x = np.matrix(x) if x.shape[0] == 1: return x.T return x class PR2Arms(object): def __init__(self, primary_ft_sensor): log("Loading PR2Arms") self.arms = PR2Arms_kdl() # KDL chain. self.joint_names_list = [['r_shoulder_pan_joint', 'r_shoulder_lift_joint', 'r_upper_arm_roll_joint', 'r_elbow_flex_joint', 'r_forearm_roll_joint', 'r_wrist_flex_joint', 'r_wrist_roll_joint'], ['l_shoulder_pan_joint', 'l_shoulder_lift_joint', 'l_upper_arm_roll_joint', 'l_elbow_flex_joint', 'l_forearm_roll_joint', 'l_wrist_flex_joint', 'l_wrist_roll_joint']] self.arm_state_lock = [RLock(), RLock()] self.jep = [None, None] self.arm_angles = [None, None] self.torso_position = None self.arm_efforts = [None, None] self.r_arm_cart_pub = rospy.Publisher('/r_cart/command_pose', PoseStamped) self.l_arm_cart_pub = rospy.Publisher('/l_cart/command_pose', PoseStamped) rospy.Subscriber('/r_cart/state', JTTeleopControllerState, self.r_cart_state_cb) rospy.Subscriber('/l_cart/state', JTTeleopControllerState, self.l_cart_state_cb) rospy.Subscriber('/joint_states', JointState, self.joint_states_cb, queue_size=2) self.marker_pub = rospy.Publisher('/pr2_arms/viz_markers', Marker) self.cep_marker_id = 1 self.r_arm_ftc = ftc.FTClient('force_torque_ft2') self.r_arm_ftc_estimate = ftc.FTClient('force_torque_ft2_estimate') self.tf_lstnr = tf.TransformListener() if primary_ft_sensor == 'ati': self.get_wrist_force = self.get_wrist_force_ati if primary_ft_sensor == 'estimate': self.get_wrist_force = self.get_wrist_force_estimate r_action_client = actionlib.SimpleActionClient('r_arm_controller/joint_trajectory_action', JointTrajectoryAction) l_action_client = actionlib.SimpleActionClient('l_arm_controller/joint_trajectory_action', JointTrajectoryAction) self.joint_action_client = [r_action_client, l_action_client] r_gripper_client = actionlib.SimpleActionClient('r_gripper_controller/gripper_action', Pr2GripperCommandAction) l_gripper_client = actionlib.SimpleActionClient('l_gripper_controller/gripper_action', Pr2GripperCommandAction) self.gripper_action_client = [r_gripper_client, l_gripper_client] rospy.sleep(2.) # self.joint_action_client[0].wait_for_server() # self.joint_action_client[1].wait_for_server() self.gripper_action_client[0].wait_for_server() self.gripper_action_client[1].wait_for_server() log("Finished loading SimpleArmManger") ## # Callback for /joint_states topic. Updates current joint # angles and efforts for the arms constantly # @param data JointState message recieved from the /joint_states topic def joint_states_cb(self, data): arm_angles = [[], []] arm_efforts = [[], []] r_jt_idx_list = [0]*7 l_jt_idx_list = [0]*7 for i, jt_nm in enumerate(self.joint_names_list[0]): r_jt_idx_list[i] = data.name.index(jt_nm) for i, jt_nm in enumerate(self.joint_names_list[1]): l_jt_idx_list[i] = data.name.index(jt_nm) for i in range(7): idx = r_jt_idx_list[i] if data.name[idx] != self.joint_names_list[0][i]: raise RuntimeError('joint angle name does not match. Expected: %s, Actual: %s i: %d'%('r_'+nm+'_joint', data.name[idx], i)) arm_angles[0].append(data.position[idx]) arm_efforts[0].append(data.effort[idx]) idx = l_jt_idx_list[i] if data.name[idx] != self.joint_names_list[1][i]: raise RuntimeError('joint angle name does not match. Expected: %s, Actual: %s i: %d'%('r_'+nm+'_joint', data.name[idx], i)) #ang = tr.angle_within_mod180(data.position[idx]) ang = data.position[idx] arm_angles[1] += [ang] arm_efforts[1] += [data.effort[idx]] self.arm_state_lock[0].acquire() self.arm_angles[0] = arm_angles[0] self.arm_efforts[0] = arm_efforts[0] torso_idx = data.name.index('torso_lift_joint') self.torso_position = data.position[torso_idx] self.arm_state_lock[0].release() self.arm_state_lock[1].acquire() self.arm_angles[1] = arm_angles[1] self.arm_efforts[1] = arm_efforts[1] self.arm_state_lock[1].release() def r_cart_state_cb(self, msg): trans, quat = self.tf_lstnr.lookupTransform('/torso_lift_link', 'r_gripper_tool_frame', rospy.Time(0)) rot = tr.quaternion_to_matrix(quat) tip = np.matrix([0.12, 0., 0.]).T self.r_ee_pos = rot*tip + np.matrix(trans).T self.r_ee_rot = rot marker = Marker() marker.header.frame_id = 'torso_lift_link' time_stamp = rospy.Time.now() marker.header.stamp = time_stamp marker.ns = 'aloha land' marker.type = Marker.SPHERE marker.action = Marker.ADD marker.pose.position.x = self.r_ee_pos[0,0] marker.pose.position.y = self.r_ee_pos[1,0] marker.pose.position.z = self.r_ee_pos[2,0] size = 0.02 marker.scale.x = size marker.scale.y = size marker.scale.z = size marker.lifetime = rospy.Duration() marker.id = 71 marker.pose.orientation.x = 0 marker.pose.orientation.y = 0 marker.pose.orientation.z = 0 marker.pose.orientation.w = 1 color = (0.5, 0., 0.0) marker.color.r = color[0] marker.color.g = color[1] marker.color.b = color[2] marker.color.a = 1. self.marker_pub.publish(marker) ros_pt = msg.x_desi_filtered.pose.position x, y, z = ros_pt.x, ros_pt.y, ros_pt.z self.r_cep_pos = np.matrix([x, y, z]).T pt = rot.T * (np.matrix([x,y,z]).T - np.matrix(trans).T) pt = pt + tip self.r_cep_pos_hooktip = rot*pt + np.matrix(trans).T ros_quat = msg.x_desi_filtered.pose.orientation quat = (ros_quat.x, ros_quat.y, ros_quat.z, ros_quat.w) self.r_cep_rot = tr.quaternion_to_matrix(quat) def l_cart_state_cb(self, msg): ros_pt = msg.x_desi_filtered.pose.position self.l_cep_pos = np.matrix([ros_pt.x, ros_pt.y, ros_pt.z]).T ros_quat = msg.x_desi_filtered.pose.orientation quat = (ros_quat.x, ros_quat.y, ros_quat.z, ros_quat.w) self.l_cep_rot = tr.quaternion_to_matrix(quat) ## Returns the current position, rotation of the arm. # @param arm 0 for right, 1 for left # @return rotation, position def end_effector_pos(self, arm): q = self.get_joint_angles(arm) return self.arms.FK_all(arm, q) ## Returns the list of 7 joint angles # @param arm 0 for right, 1 for left # @return list of 7 joint angles def get_joint_angles(self, arm): if arm != 1: arm = 0 self.arm_state_lock[arm].acquire() q = self.arm_angles[arm] self.arm_state_lock[arm].release() return q def set_jep(self, arm, q, duration=0.15): self.arm_state_lock[arm].acquire() jtg = JointTrajectoryGoal() jtg.trajectory.joint_names = self.joint_names_list[arm] jtp = JointTrajectoryPoint() jtp.positions = q #jtp.velocities = [0 for i in range(len(q))] #jtp.accelerations = [0 for i in range(len(q))] jtp.time_from_start = rospy.Duration(duration) jtg.trajectory.points.append(jtp) self.joint_action_client[arm].send_goal(jtg) self.jep[arm] = q cep, r = self.arms.FK_all(arm, q) self.arm_state_lock[arm].release() o = np.matrix([0.,0.,0.,1.]).T cep_marker = hv.single_marker(cep, o, 'sphere', '/torso_lift_link', color=(0., 0., 1., 1.), scale = (0.02, 0.02, 0.02), m_id = self.cep_marker_id) cep_marker.header.stamp = rospy.Time.now() self.marker_pub.publish(cep_marker) def get_jep(self, arm): self.arm_state_lock[arm].acquire() jep = copy.copy(self.jep[arm]) self.arm_state_lock[arm].release() return jep def get_ee_jtt(self, arm): if arm == 0: return self.r_ee_pos, self.r_ee_rot else: return self.l_ee_pos, self.l_ee_rot def get_cep_jtt(self, arm, hook_tip = False): if arm == 0: if hook_tip: return self.r_cep_pos_hooktip, self.r_cep_rot else: return self.r_cep_pos, self.r_cep_rot else: return self.l_cep_pos, self.l_cep_rot # set a cep using the Jacobian Transpose controller. def set_cep_jtt(self, arm, p, rot=None): if arm != 1: arm = 0 ps = PoseStamped() ps.header.stamp = rospy.rostime.get_rostime() ps.header.frame_id = 'torso_lift_link' ps.pose.position.x = p[0,0] ps.pose.position.y = p[1,0] ps.pose.position.z = p[2,0] if rot == None: if arm == 0: rot = self.r_cep_rot else: rot = self.l_cep_rot quat = tr.matrix_to_quaternion(rot) ps.pose.orientation.x = quat[0] ps.pose.orientation.y = quat[1] ps.pose.orientation.z = quat[2] ps.pose.orientation.w = quat[3] if arm == 0: self.r_arm_cart_pub.publish(ps) else: self.l_arm_cart_pub.publish(ps) # rotational interpolation unimplemented. def go_cep_jtt(self, arm, p): step_size = 0.01 sleep_time = 0.1 cep_p, cep_rot = self.get_cep_jtt(arm) unit_vec = (p-cep_p) unit_vec = unit_vec / np.linalg.norm(unit_vec) while np.linalg.norm(p-cep_p) > step_size: cep_p += unit_vec * step_size self.set_cep_jtt(arm, cep_p) rospy.sleep(sleep_time) self.set_cep_jtt(arm, p) rospy.sleep(sleep_time) #----------- forces ------------ # force that is being applied on the wrist. (estimate as returned # by the cartesian controller) def get_wrist_force_estimate(self, arm, bias = True, base_frame = False): if arm != 0: rospy.logerr('Unsupported arm: %d'%arm) raise RuntimeError('Unimplemented function') f = self.r_arm_ftc_estimate.read(without_bias = not bias) f = f[0:3, :] if base_frame: trans, quat = self.tf_lstnr.lookupTransform('/torso_lift_link', '/ft2_estimate', rospy.Time(0)) rot = tr.quaternion_to_matrix(quat) f = rot * f return -f # the negative is intentional (Advait, Nov 24. 2010.) # force that is being applied on the wrist. def get_wrist_force_ati(self, arm, bias = True, base_frame = False): if arm != 0: rospy.logerr('Unsupported arm: %d'%arm) raise RuntimeError('Unimplemented function') f = self.r_arm_ftc.read(without_bias = not bias) f = f[0:3, :] if base_frame: trans, quat = self.tf_lstnr.lookupTransform('/torso_lift_link', '/ft2', rospy.Time(0)) rot = tr.quaternion_to_matrix(quat) f = rot * f return -f # the negative is intentional (Advait, Nov 24. 2010.) ## Returns the list of 7 joint angles # @param arm 0 for right, 1 for left # @return list of 7 joint angles def get_force_from_torques(self, arm): if arm != 1: arm = 0 self.arm_state_lock[arm].acquire() q = self.arm_angles[arm] tau = self.arm_efforts[arm] self.arm_state_lock[arm].release() p, _ = self.arms.FK_all(arm, q) J = self.arms.Jacobian(arm, q, p) f = np.linalg.pinv(J.T) * np.matrix(tau).T f = f[0:3,:] return -f def bias_wrist_ft(self, arm): if arm != 0: rospy.logerr('Unsupported arm: %d'%arm) raise RuntimeError('Unimplemented function') self.r_arm_ftc.bias() self.r_arm_ftc_estimate.bias() #-------- gripper functions ------------ def move_gripper(self, arm, amount=0.08, effort = 15): self.gripper_action_client[arm].send_goal(Pr2GripperCommandGoal(Pr2GripperCommand(position=amount, max_effort = effort))) ## Open the gripper # @param arm 0 for right, 1 for left def open_gripper(self, arm): self.move_gripper(arm, 0.08, -1) ## Close the gripper # @param arm 0 for right, 1 for left def close_gripper(self, arm, effort = 15): self.move_gripper(arm, 0.0, effort) ## # using KDL for pr2 arm kinematics. class PR2Arms_kdl(): def __init__(self): self.right_chain = self.create_right_chain() fk, ik_v, ik_p, jac = self.create_solvers(self.right_chain) self.right_fk = fk self.right_ik_v = ik_v self.right_ik_p = ik_p self.right_jac = jac self.right_tooltip = np.matrix([0.,0.,0.]).T def create_right_chain(self): ch = kdl.Chain() self.right_arm_base_offset_from_torso_lift_link = np.matrix([0., -0.188, 0.]).T # shoulder pan ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotZ),kdl.Frame(kdl.Vector(0.1,0.,0.)))) # shoulder lift ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotY),kdl.Frame(kdl.Vector(0.,0.,0.)))) # upper arm roll ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotX),kdl.Frame(kdl.Vector(0.4,0.,0.)))) # elbox flex ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotY),kdl.Frame(kdl.Vector(0.0,0.,0.)))) # forearm roll ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotX),kdl.Frame(kdl.Vector(0.321,0.,0.)))) # wrist flex ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotY),kdl.Frame(kdl.Vector(0.,0.,0.)))) # wrist roll ch.addSegment(kdl.Segment(kdl.Joint(kdl.Joint.RotX),kdl.Frame(kdl.Vector(0.,0.,0.)))) return ch def create_solvers(self, ch): fk = kdl.ChainFkSolverPos_recursive(ch) ik_v = kdl.ChainIkSolverVel_pinv(ch) ik_p = kdl.ChainIkSolverPos_NR(ch, fk, ik_v) jac = kdl.ChainJntToJacSolver(ch) return fk, ik_v, ik_p, jac ## define tooltip as a 3x1 np matrix in the wrist coord frame. def set_tooltip(self, arm, p): if arm == 0: self.right_tooltip = p else: rospy.logerr('Arm %d is not supported.'%(arm)) def FK_kdl(self, arm, q, link_number): if arm == 0: fk = self.right_fk endeffec_frame = kdl.Frame() kinematics_status = fk.JntToCart(q, endeffec_frame, link_number) if kinematics_status >= 0: return endeffec_frame else: rospy.loginfo('Could not compute forward kinematics.') return None else: msg = '%s arm not supported.'%(arm) rospy.logerr(msg) raise RuntimeError(msg) ## returns point in torso lift link. def FK_all(self, arm, q, link_number = 7): q = self.pr2_to_kdl(q) frame = self.FK_kdl(arm, q, link_number) pos = frame.p pos = ku.kdl_vec_to_np(pos) pos = pos + self.right_arm_base_offset_from_torso_lift_link m = frame.M rot = ku.kdl_rot_to_np(m) if arm == 0: tooltip_baseframe = rot * self.right_tooltip pos += tooltip_baseframe else: rospy.logerr('Arm %d is not supported.'%(arm)) return None return pos, rot def kdl_to_pr2(self, q): if q == None: return None q_pr2 = [0] * 7 q_pr2[0] = q[0] q_pr2[1] = q[1] q_pr2[2] = q[2] q_pr2[3] = q[3] q_pr2[4] = q[4] q_pr2[5] = q[5] q_pr2[6] = q[6] return q_pr2 def pr2_to_kdl(self, q): if q == None: return None n = len(q) q_kdl = kdl.JntArray(n) for i in range(n): q_kdl[i] = q[i] return q_kdl def Jac_kdl(self, arm, q): J_kdl = kdl.Jacobian(7) if arm != 0: rospy.logerr('Unsupported arm: '+ str(arm)) return None self.right_jac.JntToJac(q,J_kdl) kdl_jac = np.matrix([ [J_kdl[0,0],J_kdl[0,1],J_kdl[0,2],J_kdl[0,3],J_kdl[0,4],J_kdl[0,5],J_kdl[0,6]], [J_kdl[1,0],J_kdl[1,1],J_kdl[1,2],J_kdl[1,3],J_kdl[1,4],J_kdl[1,5],J_kdl[1,6]], [J_kdl[2,0],J_kdl[2,1],J_kdl[2,2],J_kdl[2,3],J_kdl[2,4],J_kdl[2,5],J_kdl[2,6]], [J_kdl[3,0],J_kdl[3,1],J_kdl[3,2],J_kdl[3,3],J_kdl[3,4],J_kdl[3,5],J_kdl[3,6]], [J_kdl[4,0],J_kdl[4,1],J_kdl[4,2],J_kdl[4,3],J_kdl[4,4],J_kdl[4,5],J_kdl[4,6]], [J_kdl[5,0],J_kdl[5,1],J_kdl[5,2],J_kdl[5,3],J_kdl[5,4],J_kdl[5,5],J_kdl[5,6]], ]) return kdl_jac # ## compute Jacobian (at wrist). # # @param arm - 0 or 1 # # @param q - list of 7 joint angles. # # @return 6x7 np matrix # def Jac(self, arm, q): # rospy.logerr('Jac only works for getting the Jacobian at the wrist. Use Jacobian to get the Jacobian at a general location.') # jntarr = self.pr2_to_kdl(q) # kdl_jac = self.Jac_kdl(arm, jntarr) # pr2_jac = kdl_jac # return pr2_jac ## compute Jacobian at point pos. # p is in the torso_lift_link coord frame. def Jacobian(self, arm, q, pos): if arm != 0: rospy.logerr('Arm %d is not supported.'%(arm)) return None tooltip = self.right_tooltip self.right_tooltip = np.matrix([0.,0.,0.]).T v_list = [] w_list = [] for i in range(7): p, rot = self.FK_all(arm, q, i) r = pos - p z_idx = self.right_chain.getSegment(i).getJoint().getType() - 1 z = rot[:, z_idx] v_list.append(np.matrix(np.cross(z.A1, r.A1)).T) w_list.append(z) J = np.row_stack((np.column_stack(v_list), np.column_stack(w_list))) self.right_tooltip = tooltip return J def close_to_singularity(self, arm, q): pass def within_joint_limits(self, arm, q, delta_list=[0., 0., 0., 0., 0., 0., 0.]): if arm == 0: # right arm min_arr = np.radians(np.array([-109., -24, -220, -132, -np.inf, -120, -np.inf])) #max_arr = np.radians(np.array([26., 68, 41, 0, np.inf, 0, np.inf])) max_arr = np.radians(np.array([26., 68, 41, 5, np.inf, 5, np.inf])) # 5 to prevent singularity. Need to come up with a better solution. else: raise RuntimeError('within_joint_limits unimplemented for left arm') q_arr = np.array(q) d_arr = np.array(delta_list) return np.all((q_arr <= max_arr+d_arr, q_arr >= min_arr+d_arr)) if __name__ == '__main__': from visualization_msgs.msg import Marker import hrl_lib.viz as hv rospy.init_node('pr2_arms_test') pr2_arms = PR2Arms() pr2_kdl = PR2Arms_kdl() r_arm, l_arm = 0, 1 arm = r_arm if True: np.set_printoptions(precision=2, suppress=True) while not rospy.is_shutdown(): q = pr2_arms.get_joint_angles(arm) print 'q in degrees:', np.degrees(q) rospy.sleep(0.1) if False: jep = [0.] * 7 rospy.loginfo('Going to home location.') raw_input('Hit ENTER to go') pr2_arms.set_jep(arm, jep, duration=2.) if False: # testing FK by publishing a frame marker. marker_pub = rospy.Publisher('/pr2_kdl/ee_marker', Marker) pr2_kdl.set_tooltip(arm, np.matrix([0.15, 0., 0.]).T) rt = rospy.Rate(100) rospy.loginfo('Starting the maker publishing loop.') while not rospy.is_shutdown(): q = pr2_arms.get_joint_angles(arm) p, rot = pr2_kdl.FK_all(arm, q) m = hv.create_frame_marker(p, rot, 0.15, '/torso_lift_link') m.header.stamp = rospy.Time.now() marker_pub.publish(m) rt.sleep() if False: # testing Jacobian by printing KDL and my own Jacobian at the # current configuration. while not rospy.is_shutdown(): q = pr2_arms.get_joint_angles(arm) J_kdl = pr2_kdl.Jac(arm , q) p, rot = pr2_kdl.FK_all(arm, q) J_adv = pr2_kdl.Jacobian(arm, q, p) print J_adv.shape diff_J = J_kdl - J_adv print 'difference between KDL and adv is:' print diff_J print 'Norm of difference matrix:', np.linalg.norm(diff_J) raw_input('Move arm into some configuration and hit enter to get the Jacobian.') # ## Performs Inverse Kinematics on the given position and rotation # # @param arm 0 for right, 1 for left # # @param p cartesian position in torso_lift_link frame # # @param rot quaternion rotation column or rotation matrix # # of wrist in torso_lift_link frame # # @param q_guess initial joint angles to use for finding IK # def IK(self, arm, p, rot, q_guess): # if arm != 1: # arm = 0 # # p = make_column(p) # # if rot.shape == (3, 3): # quat = np.matrix(tr.matrix_to_quaternion(rot)).T # elif rot.shape == (4, 1): # quat = make_column(rot) # else: # rospy.logerr('Inverse kinematics failed (bad rotation)') # return None # # ik_req = GetPositionIKRequest() # ik_req.timeout = rospy.Duration(5.) # if arm == 0: # ik_req.ik_request.ik_link_name = 'r_wrist_roll_link' # else: # ik_req.ik_request.ik_link_name = 'l_wrist_roll_link' # ik_req.ik_request.pose_stamped.header.frame_id = 'torso_lift_link' # # ik_req.ik_request.pose_stamped.pose.position.x = p[0,0] # ik_req.ik_request.pose_stamped.pose.position.y = p[1,0] # ik_req.ik_request.pose_stamped.pose.position.z = p[2,0] # # ik_req.ik_request.pose_stamped.pose.orientation.x = quat[0] # ik_req.ik_request.pose_stamped.pose.orientation.y = quat[1] # ik_req.ik_request.pose_stamped.pose.orientation.z = quat[2] # ik_req.ik_request.pose_stamped.pose.orientation.w = quat[3] # # ik_req.ik_request.ik_seed_state.joint_state.position = q_guess # ik_req.ik_request.ik_seed_state.joint_state.name = self.joint_names_list[arm] # # ik_resp = self.ik_srv[arm].call(ik_req) # if ik_resp.error_code.val == ik_resp.error_code.SUCCESS: # ret = list(ik_resp.solution.joint_state.position) # else: # rospy.logerr('Inverse kinematics failed') # ret = None # # return ret #
[ [ 1, 0, 0.0043, 0.0014, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0058, 0.0014, 0, 0.66, 0.0323, 83, 0, 2, 0, 0, 83, 0, 0 ], [ 1, 0, 0.0072, 0.0014, 0, 0....
[ "import numpy as np, math", "from threading import RLock, Timer", "import sys, copy", "import roslib; roslib.load_manifest('hrl_pr2_lib')", "import roslib; roslib.load_manifest('hrl_pr2_lib')", "roslib.load_manifest('force_torque') # hack by Advait", "import force_torque.FTClient as ftc", "import tf",...
# # Temoprarily in this package. Advait needs to move it to a better # location. # import numpy as np, math import copy import roslib; roslib.load_manifest('hrl_pr2_door_opening') import rospy import hrl_lib.util as ut ## Class defining the core EPC function and a few simple examples. # More complex behaviors that use EPC should have their own ROS # packages. class EPC(): def __init__(self, robot): self.robot = robot self.f_list = [] self.ee_list = [] self.cep_list = [] ## # @param equi_pt_generator: function that returns stop, ea where ea: equilibrium angles and stop: string which is '' for epc motion to continue # @param rapid_call_func: called in the time between calls to the equi_pt_generator can be used for logging, safety etc. returns string which is '' for epc motion to continue # @param time_step: time between successive calls to equi_pt_generator # @param arg_list - list of arguments to be passed to the equi_pt_generator # @return stop (the string which has the reason why the epc # motion stopped.), ea (last commanded equilibrium angles) def epc_motion(self, equi_pt_generator, time_step, arm, arg_list, rapid_call_func=None, control_function=None): stop, ea = equi_pt_generator(*arg_list) t_end = rospy.get_time() while stop == '': if rospy.is_shutdown(): stop = 'rospy shutdown' continue t_end += time_step #self.robot.set_jointangles(arm, ea) #import pdb; pdb.set_trace() control_function(arm, *ea) # self.robot.step() this should be within the rapid_call_func for the meka arms. t1 = rospy.get_time() while t1<t_end: if rapid_call_func != None: stop = rapid_call_func(arm) if stop != '': break #self.robot.step() this should be within the rapid_call_func for the meka arms rospy.sleep(0.01) t1 = rospy.get_time() if stop == '': stop, ea = equi_pt_generator(*arg_list) if stop == 'reset timing': stop = '' t_end = rospy.get_time() return stop, ea ## Pull back along a straight line (-ve x direction) # @param arm - 'right_arm' or 'left_arm' # @param distance - how far back to pull. def pull_back_cartesian_control(self, arm, distance, logging_fn): cep, _ = self.robot.get_cep_jtt(arm) cep_end = cep + distance * np.matrix([-1., 0., 0.]).T self.dist_left = distance def eq_gen_pull_back(cep): logging_fn(arm) if self.dist_left <= 0.: return 'done', None step_size = 0.01 cep[0,0] -= step_size self.dist_left -= step_size if cep[0,0] < 0.4: return 'very close to the body: %.3f'%cep[0,0], None return '', (cep, None) arg_list = [cep] s = self.epc_motion(eq_gen_pull_back, 0.1, arm, arg_list, control_function = self.robot.set_cep_jtt) print s def move_till_hit(self, arm, vec=np.matrix([0.3,0.,0.]).T, force_threshold=3.0, speed=0.1, bias_FT=True): unit_vec = vec/np.linalg.norm(vec) time_step = 0.1 dist = np.linalg.norm(vec) step_size = speed * time_step cep_start, _ = self.robot.get_cep_jtt(arm) cep = copy.copy(cep_start) def eq_gen(cep): force = self.robot.get_wrist_force(arm, base_frame = True) force_projection = force.T*unit_vec *-1 # projection in direction opposite to motion. print 'force_projection:', force_projection if force_projection>force_threshold: return 'done', None elif np.linalg.norm(force)>45.: return 'large force', None elif np.linalg.norm(cep_start-cep) >= dist: return 'reached without contact', None else: cep_t = cep + unit_vec * step_size cep[0,0] = cep_t[0,0] cep[1,0] = cep_t[1,0] cep[2,0] = cep_t[2,0] return '', (cep, None) if bias_FT: self.robot.bias_wrist_ft(arm) rospy.sleep(0.5) return self.epc_motion(eq_gen, time_step, arm, [cep], control_function = self.robot.set_cep_jtt) def cep_gen_surface_follow(self, arm, move_dir, force_threshold, cep, cep_start): wrist_force = self.robot.get_wrist_force(arm, base_frame=True) if wrist_force[0,0] < -3.: cep[0,0] -= 0.002 if wrist_force[0,0] > -1.: cep[0,0] += 0.003 if cep[0,0] > (cep_start[0,0]+0.05): cep[0,0] = cep_start[0,0]+0.05 step_size = 0.002 cep_t = cep + move_dir * step_size cep[0,0] = cep_t[0,0] cep[1,0] = cep_t[1,0] cep[2,0] = cep_t[2,0] print 'wrist_force:', wrist_force.A1 v = cep - cep_start if (wrist_force.T * move_dir)[0,0] < -force_threshold: stop = 'got a hook' elif np.linalg.norm(wrist_force) > 50.: stop = 'force is large %f'%(np.linalg.norm(wrist_force)) elif (v.T * move_dir)[0,0] > 0.20: stop = 'moved a lot without a hook' else: stop = '' return stop, (cep, None) if __name__ == '__main__': import pr2_arms as pa rospy.init_node('epc_pr2', anonymous = True) rospy.logout('epc_pr2: ready') pr2_arms = pa.PR2Arms() epc = EPC(pr2_arms) r_arm, l_arm = 0, 1 arm = r_arm # #----- testing move_till_hit ------ # p1 = np.matrix([0.6, -0.22, -0.05]).T # epc.robot.go_cep_jtt(arm, p1) # epc.move_till_hit(arm) raw_input('Hit ENTER to close') pr2_arms.close_gripper(arm) raw_input('Hit ENTER to search_and_hook') p1 = np.matrix([0.8, -0.22, -0.05]).T epc.search_and_hook(arm, p1) epc.pull_back_cartesian_control(arm, 0.3) d = { 'f_list': epc.f_list, 'ee_list': epc.ee_list, 'cep_list': epc.cep_list } ut.save_pickle(d, 'pr2_pull_'+ut.formatted_time()+'.pkl') # if False: # ea = [0, 0, 0, 0, 0, 0, 0] # ea = epc.robot.get_joint_angles(arm) # rospy.logout('Going to starting position') # epc.robot.set_jointangles(arm, ea, duration=4.0) # raw_input('Hit ENTER to pull') # epc.pull_back(arm, ea, tr.Rx(0), 0.2) # # if False: # p = np.matrix([0.9, -0.3, -0.15]).T # rot = tr.Rx(0.) # rot = tr.Rx(math.radians(90.)) # # rospy.logout('Going to starting position') # # epc.robot.open_gripper(arm) # epc.robot.set_cartesian(arm, p, rot) # # raw_input('Hit ENTER to close the gripper') # # epc.robot.close_gripper(arm) # raw_input('Hit ENTER to pull') # epc.pull_back_cartesian_control(arm, p, rot, 0.4)
[ [ 1, 0, 0.0341, 0.0049, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.039, 0.0049, 0, 0.66, 0.1429, 739, 0, 1, 0, 0, 739, 0, 0 ], [ 1, 0, 0.0488, 0.0049, 0, 0...
[ "import numpy as np, math", "import copy", "import roslib; roslib.load_manifest('hrl_pr2_door_opening')", "import roslib; roslib.load_manifest('hrl_pr2_door_opening')", "import rospy", "import hrl_lib.util as ut", "class EPC():\n def __init__(self, robot):\n self.robot = robot\n self.f_...
import numpy as np, math import copy from threading import RLock import roslib; roslib.load_manifest('hrl_pr2_door_opening') roslib.load_manifest('equilibrium_point_control') import rospy from equilibrium_point_control.msg import MechanismKinematicsRot from equilibrium_point_control.msg import MechanismKinematicsJac from equilibrium_point_control.msg import ForceTrajectory from geometry_msgs.msg import Point32 from std_msgs.msg import Empty import epc import hrl_lib.util as ut class Door_EPC(epc.EPC): def __init__(self, robot): epc.EPC.__init__(self, robot) self.mech_kinematics_lock = RLock() self.fit_circle_lock = RLock() rospy.Subscriber('mechanism_kinematics_rot', MechanismKinematicsRot, self.mechanism_kinematics_rot_cb) rospy.Subscriber('epc/stop', Empty, self.stop_cb) # used in the ROS stop_cb and equi_pt_generator_control_radial_force self.force_traj_pub = rospy.Publisher('epc/force_test', ForceTrajectory) self.mech_traj_pub = rospy.Publisher('mechanism_trajectory', Point32) def init_log(self): self.f_list = [] self.f_list_ati = [] self.f_list_estimate = [] self.f_list_torques = [] self.cep_list = [] self.ee_list = [] self.ft = ForceTrajectory() if self.mechanism_type != '': self.ft.type = self.mechanism_type else: self.ft.type = 'rotary' def log_state(self, arm): # only logging the right arm. f = self.robot.get_wrist_force_ati(arm, base_frame=True) self.f_list_ati.append(f.A1.tolist()) f = self.robot.get_wrist_force_estimate(arm, base_frame=True) self.f_list_estimate.append(f.A1.tolist()) f = self.robot.get_force_from_torques(arm) self.f_list_torques.append(f.A1.tolist()) f = self.robot.get_wrist_force(arm, base_frame=True) self.f_list.append(f.A1.tolist()) cep, _ = self.robot.get_cep_jtt(arm, hook_tip=True) self.cep_list.append(cep.A1.tolist()) # ee, _ = self.robot.get_ee_jtt(arm) ee, _ = self.robot.end_effector_pos(arm) self.ee_list.append(ee.A1.tolist()) if self.started_pulling_on_handle == False: if f[0,0] > 10.: self.started_pulling_on_handle_count += 1 else: self.started_pulling_on_handle_count = 0 self.init_log() # reset logs until started pulling on the handle. self.init_tangent_vector = None if self.started_pulling_on_handle_count > 1: self.started_pulling_on_handle = True return '' ## ROS callback. Stop and maintain position. def stop_cb(self, cmd): self.stopping_string = 'stop_cb called.' def common_stopping_conditions(self): stop = '' # right arm only. wrist_force = self.robot.get_wrist_force(0, base_frame=True) mag = np.linalg.norm(wrist_force) if mag > self.eq_force_threshold: stop = 'force exceed' if mag < 1.2 and self.hooked_location_moved: if (self.prev_force_mag - mag) > 30.: stop = 'slip: force step decrease and below thresold.' else: self.slip_count += 1 else: self.slip_count = 0 if self.slip_count == 10: stop = 'slip: force below threshold for too long.' return stop def mechanism_kinematics_rot_cb(self, mk): self.fit_circle_lock.acquire() self.cx_start = mk.cx self.cy_start = mk.cy self.cz_start = mk.cz self.rad = mk.rad self.fit_circle_lock.release() ## constantly update the estimate of the kinematics and move the # equilibrium point along the tangent of the estimated arc, and # try to keep the radial force constant. # @param h_force_possible - True (hook side) or False (hook up). # @param v_force_possible - False (hook side) or True (hook up). # Is maintaining a radial force possible or not (based on hook # geometry and orientation) # @param cep_vel - tangential velocity of the cep in m/s def cep_gen_control_radial_force(self, arm, cep, cep_vel): self.log_state(arm) if self.started_pulling_on_handle == False: cep_vel = 0.02 #step_size = 0.01 * cep_vel step_size = 0.1 * cep_vel # 0.1 is the time interval between calls to the equi_generator function (see pull) stop = self.common_stopping_conditions() wrist_force = self.robot.get_wrist_force(arm, base_frame=True) mag = np.linalg.norm(wrist_force) curr_pos, _ = self.robot.get_ee_jtt(arm) if len(self.ee_list)>1: start_pos = np.matrix(self.ee_list[0]).T else: start_pos = curr_pos #mechanism kinematics. if self.started_pulling_on_handle: self.mech_traj_pub.publish(Point32(curr_pos[0,0], curr_pos[1,0], curr_pos[2,0])) self.fit_circle_lock.acquire() rad = self.rad cx_start, cy_start = self.cx_start, self.cy_start cz_start = self.cz_start self.fit_circle_lock.release() cx, cy = cx_start, cy_start cz = cz_start print 'cx, cy, r:', cx, cy, rad radial_vec = curr_pos - np.matrix([cx,cy,cz]).T radial_vec = radial_vec/np.linalg.norm(radial_vec) if cy_start < start_pos[1,0]: tan_x,tan_y = -radial_vec[1,0],radial_vec[0,0] else: tan_x,tan_y = radial_vec[1,0],-radial_vec[0,0] if tan_x > 0. and (start_pos[0,0]-curr_pos[0,0]) < 0.09: tan_x = -tan_x tan_y = -tan_y if cy_start > start_pos[1,0]: radial_vec = -radial_vec # axis to the left, want force in # anti-radial direction. rv = radial_vec force_vec = np.matrix([rv[0,0], rv[1,0], 0.]).T tangential_vec = np.matrix([tan_x, tan_y, 0.]).T tangential_vec_ts = tangential_vec radial_vec_ts = radial_vec force_vec_ts = force_vec if arm == 'right_arm' or arm == 0: if force_vec_ts[1,0] < 0.: # only allowing force to the left force_vec_ts = -force_vec_ts else: if force_vec_ts[1,0] > 0.: # only allowing force to the right force_vec_ts = -force_vec_ts f_vec = -1*np.array([wrist_force[0,0], wrist_force[1,0], wrist_force[2,0]]) f_rad_mag = np.dot(f_vec, force_vec.A1) err = f_rad_mag-4. if err>0.: kp = -0.1 else: kp = -0.2 radial_motion_mag = kp * err # radial_motion_mag in cm (depends on eq_motion step size) radial_motion_vec = force_vec * radial_motion_mag print 'tangential_vec:', tangential_vec.A1 eq_motion_vec = copy.copy(tangential_vec) eq_motion_vec += radial_motion_vec self.prev_force_mag = mag if self.init_tangent_vector == None or self.started_pulling_on_handle == False: self.init_tangent_vector = copy.copy(tangential_vec_ts) c = np.dot(tangential_vec_ts.A1, self.init_tangent_vector.A1) ang = np.arccos(c) if np.isnan(ang): ang = 0. tangential_vec = tangential_vec / np.linalg.norm(tangential_vec) # paranoia abot vectors not being unit vectors. dist_moved = np.dot((curr_pos - start_pos).A1, tangential_vec_ts.A1) ftan = abs(np.dot(wrist_force.A1, tangential_vec.A1)) self.ft.tangential_force.append(ftan) self.ft.radial_force.append(f_rad_mag) if self.ft.type == 'rotary': self.ft.configuration.append(ang) else: # drawer print 'dist_moved:', dist_moved self.ft.configuration.append(dist_moved) if self.started_pulling_on_handle: self.force_traj_pub.publish(self.ft) # if self.started_pulling_on_handle == False: # ftan_pull_test = -np.dot(wrist_force.A1, tangential_vec.A1) # print 'ftan_pull_test:', ftan_pull_test # if ftan_pull_test > 5.: # self.started_pulling_on_handle_count += 1 # else: # self.started_pulling_on_handle_count = 0 # self.init_log() # reset logs until started pulling on the handle. # self.init_tangent_vector = None # # if self.started_pulling_on_handle_count > 1: # self.started_pulling_on_handle = True if abs(dist_moved) > 0.09 and self.hooked_location_moved == False: # change the force threshold once the hook has started pulling. self.hooked_location_moved = True self.eq_force_threshold = ut.bound(mag+30.,20.,80.) self.ftan_threshold = 1.2 * self.ftan_threshold + 20. if self.hooked_location_moved: if abs(tangential_vec_ts[2,0]) < 0.2 and ftan > self.ftan_threshold: stop = 'ftan threshold exceed: %f'%ftan else: self.ftan_threshold = max(self.ftan_threshold, ftan) if self.hooked_location_moved and ang > math.radians(90.): print 'Angle:', math.degrees(ang) self.open_ang_exceed_count += 1 if self.open_ang_exceed_count > 2: stop = 'opened mechanism through large angle: %.1f'%(math.degrees(ang)) else: self.open_ang_exceed_count = 0 cep_t = cep + eq_motion_vec * step_size cep[0,0] = cep_t[0,0] cep[1,0] = cep_t[1,0] cep[2,0] = cep_t[2,0] print 'CEP:', cep.A1 stop = stop + self.stopping_string return stop, (cep, None) def pull(self, arm, force_threshold, cep_vel, mechanism_type=''): self.mechanism_type = mechanism_type self.stopping_string = '' self.eq_pt_not_moving_counter = 0 self.init_log() self.init_tangent_vector = None self.open_ang_exceed_count = 0. self.eq_force_threshold = force_threshold self.ftan_threshold = 2. self.hooked_location_moved = False # flag to indicate when the hooking location started moving. self.prev_force_mag = np.linalg.norm(self.robot.get_wrist_force(arm)) self.slip_count = 0 self.started_pulling_on_handle = False self.started_pulling_on_handle_count = 0 ee_pos, _ = self.robot.get_ee_jtt(arm) self.cx_start = ee_pos[0,0] self.rad = 10.0 self.cy_start = ee_pos[1,0]-self.rad self.cz_start = ee_pos[2,0] cep, _ = self.robot.get_cep_jtt(arm) arg_list = [arm, cep, cep_vel] result, _ = self.epc_motion(self.cep_gen_control_radial_force, 0.1, arm, arg_list, self.log_state, #0.01, arm, arg_list, control_function = self.robot.set_cep_jtt) print 'EPC motion result:', result print 'Original force threshold:', force_threshold print 'Adapted force threshold:', self.eq_force_threshold print 'Adapted ftan threshold:', self.ftan_threshold d = { 'f_list': self.f_list, 'ee_list': self.ee_list, 'cep_list': self.cep_list, 'ftan_list': self.ft.tangential_force, 'config_list': self.ft.configuration, 'frad_list': self.ft.radial_force, 'f_list_ati': self.f_list_ati, 'f_list_estimate': self.f_list_estimate, 'f_list_torques': self.f_list_torques } ut.save_pickle(d,'pr2_pull_'+ut.formatted_time()+'.pkl') def search_and_hook(self, arm, hook_loc, hooking_force_threshold = 5., hit_threshold=15., hit_motions = 1, hook_direction = 'left'): # this needs to be debugged. Hardcoded for now. #if arm == 'right_arm' or arm == 0: # hook_dir = np.matrix([0., 1., 0.]).T # hook direc in home position # offset = -0.03 #elif arm == 'left_arm' or arm == 1: # hook_dir = np.matrix([0., -1., 0.]).T # hook direc in home position # offset = -0.03 #else: # raise RuntimeError('Unknown arm: %s', arm) #start_loc = hook_loc + rot_mat.T * hook_dir * offset if hook_direction == 'left': offset = np.matrix([0., -0.03, 0.]).T move_dir = np.matrix([0., 1., 0.]).T elif hook_direction == 'up': offset = np.matrix([0., 0., -0.03]).T move_dir = np.matrix([0., 0., 1.]).T start_loc = hook_loc + offset # vector normal to surface and pointing into the surface. normal_tl = np.matrix([1.0, 0., 0.]).T pt1 = start_loc - normal_tl * 0.1 self.robot.go_cep_jtt(arm, pt1) raw_input('Hit ENTER to go') vec = normal_tl * 0.2 rospy.sleep(1.) for i in range(hit_motions): s = self.move_till_hit(arm, vec=vec, force_threshold=hit_threshold, speed=0.07) cep_start, _ = self.robot.get_cep_jtt(arm) cep = copy.copy(cep_start) arg_list = [arm, move_dir, hooking_force_threshold, cep, cep_start] print 'Hi there.' s = self.epc_motion(self.cep_gen_surface_follow, 0.1, arm, arg_list, control_function = self.robot.set_cep_jtt) print 'result:', s return s if __name__ == '__main__': import pr2_arms as pa rospy.init_node('epc_pr2', anonymous = True) rospy.logout('epc_pr2: ready') #pr2_arms = pa.PR2Arms(primary_ft_sensor='ati') pr2_arms = pa.PR2Arms(primary_ft_sensor='estimate') door_epc = Door_EPC(pr2_arms) r_arm, l_arm = 0, 1 arm = r_arm tip = np.matrix([0.35, 0., 0.]).T pr2_arms.arms.set_tooltip(arm, tip) raw_input('Hit ENTER to close') pr2_arms.close_gripper(arm, effort=80) raw_input('Hit ENTER to start Door Opening') # for cabinets. #p1 = np.matrix([0.8, -0.40, -0.04]).T # pos 3 #p1 = np.matrix([0.8, -0.10, -0.04]).T # pos 2 p1 = np.matrix([0.8, -0.35, 0.1]).T # pos 1 door_epc.search_and_hook(arm, p1, hook_direction='left') door_epc.pull(arm, force_threshold=40., cep_vel=0.05) # # hrl toolchest drawer. # p1 = np.matrix([0.8, -0.2, -0.17]).T # door_epc.search_and_hook(arm, p1, hook_direction='up') # door_epc.pull(arm, force_threshold=40., cep_vel=0.05)
[ [ 1, 0, 0.0052, 0.0026, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0078, 0.0026, 0, 0.66, 0.0667, 739, 0, 1, 0, 0, 739, 0, 0 ], [ 1, 0, 0.0104, 0.0026, 0, ...
[ "import numpy as np, math", "import copy", "from threading import RLock", "import roslib; roslib.load_manifest('hrl_pr2_door_opening')", "import roslib; roslib.load_manifest('hrl_pr2_door_opening')", "roslib.load_manifest('equilibrium_point_control')", "import rospy", "from equilibrium_point_control.m...
#!/usr/bin/python import numpy as np, math import copy from threading import RLock import roslib; roslib.load_manifest('hrl_pr2_door_opening') import rospy from hrl_msgs.msg import FloatArray from geometry_msgs.msg import Twist def ft_cb(data): lock.acquire() ft_val[0] = data.linear.x ft_val[1] = data.linear.y ft_val[2] = data.linear.z ft_val[3] = data.angular.x ft_val[4] = data.angular.y ft_val[5] = data.angular.z lock.release() if __name__ == '__main__': lock = RLock() ft_val = [0.] * 6 pub = rospy.Subscriber('/r_cart/state/wrench', Twist, ft_cb) pub = rospy.Publisher('/force_torque_ft2_estimate', FloatArray) rospy.init_node('ati_ft_emulator') rospy.loginfo('Started the ATI FT emulator.') rt = rospy.Rate(100) while not rospy.is_shutdown(): lock.acquire() send_ft_val = copy.copy(ft_val) lock.release() fa = FloatArray(rospy.Header(stamp=rospy.Time.now()), send_ft_val) pub.publish(fa) rt.sleep()
[ [ 1, 0, 0.0769, 0.0256, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.1026, 0.0256, 0, 0.66, 0.1111, 739, 0, 1, 0, 0, 739, 0, 0 ], [ 1, 0, 0.1282, 0.0256, 0, ...
[ "import numpy as np, math", "import copy", "from threading import RLock", "import roslib; roslib.load_manifest('hrl_pr2_door_opening')", "import roslib; roslib.load_manifest('hrl_pr2_door_opening')", "import rospy", "from hrl_msgs.msg import FloatArray", "from geometry_msgs.msg import Twist", "def f...
import roslib roslib.load_manifest('rospy') roslib.load_manifest('visualization_msgs') import rospy from visualization_msgs.msg import Marker import numpy as np class SceneDraw: def __init__(self, topic='contact_visualization', node='contact_sim', frame='/world_frame'): self.pub = rospy.Publisher(topic+'_marker', Marker) self.frame = frame self.Marker = Marker def pub_body(self, pos, quat, scale, color, num, shape, text = ''): marker = Marker() marker.header.frame_id = self.frame marker.header.stamp = rospy.Time.now() marker.ns = "basic_shapes" marker.id = num marker.type = shape marker.action = Marker.ADD marker.pose.position.x = pos[0] marker.pose.position.y = pos[1] marker.pose.position.z = pos[2] marker.pose.orientation.x = quat[0] marker.pose.orientation.y = quat[1] marker.pose.orientation.z = quat[2] marker.pose.orientation.w = quat[3] marker.scale.x = scale[0] marker.scale.y = scale[1] marker.scale.z = scale[2] marker.color.r = color[0] marker.color.g = color[1] marker.color.b = color[2] marker.color.a = color[3] marker.lifetime = rospy.Duration() marker.text = text self.pub.publish(marker) def get_rot_mat(self, rot): # rot_mat = np.matrix([[rot[0], rot[3], rot[6]],[rot[1], rot[4], rot[7]], [rot[2], rot[5], rot[8]]]) rot_mat = np.matrix([[rot[0], rot[4], rot[8]],[rot[1], rot[5], rot[9]], [rot[2], rot[6], rot[10]]]) return rot_mat
[ [ 1, 0, 0.02, 0.02, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.04, 0.02, 0, 0.66, 0.1667, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 8, 0, 0.06, 0.02, 0, 0.66, 0.33...
[ "import roslib", "roslib.load_manifest('rospy')", "roslib.load_manifest('visualization_msgs')", "import rospy", "from visualization_msgs.msg import Marker", "import numpy as np", "class SceneDraw:\n def __init__(self, topic='contact_visualization', node='contact_sim', frame='/world_frame'):\n ...
#!/usr/bin/env python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # #Author: Marc Killpack import roslib; roslib.load_manifest('pr2_playpen') import rospy from pr2_playpen.srv import * from robotis.lib_robotis import * class Play: def __init__(self): dyn = USB2Dynamixel_Device('/dev/ttyUSB0') self.playpen = Robotis_Servo(dyn, 31) self.conveyor = Robotis_Servo(dyn,32) self.conveyor.init_cont_turn() rospy.init_node('playpen_server') s_play = rospy.Service('playpen', Playpen, self.move_playpen) s_conv = rospy.Service('conveyor', Conveyor, self.move_conveyor) rospy.spin() def move_conveyor(self, data): delt_time = abs(data.distance)/2.39/0.685/0.0483*2/0.75 if data.distance > 0: self.conveyor.set_angvel(-5) else: self.conveyor.set_angvel(5) rospy.sleep(delt_time) self.conveyor.set_angvel(0) print "move conveyor" return ConveyorResponse(1) def move_playpen(self, data): if data.command == 0: self.playpen.move_angle(0.28) print "closing playpen" elif data.command == 1: self.playpen.move_angle(1.90) print "opening playpen...releasing object" return PlaypenResponse(1) if __name__== "__main__": go = Play()
[ [ 1, 0, 0.4444, 0.0139, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.4444, 0.0139, 0, 0.66, 0.1667, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.4583, 0.0139, 0, 0....
[ "import roslib; roslib.load_manifest('pr2_playpen')", "import roslib; roslib.load_manifest('pr2_playpen')", "import rospy", "from pr2_playpen.srv import *", "from robotis.lib_robotis import *", "class Play:\n\n def __init__(self):\n dyn = USB2Dynamixel_Device('/dev/ttyUSB0')\n self.plaype...
#!/usr/bin/env python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # #Author: Marc Killpack import os import roslib roslib.load_manifest('pr2_playpen') roslib.load_manifest('tf_conversions') import rospy import math import tf import sys import tf_conversions.posemath as pm import numpy as np from geometry_msgs.msg import Pose if __name__ == '__main__': rospy.init_node('playpen_calibration') listener = tf.TransformListener() trans_list = [] rot_list = [] rate = rospy.Rate(10.0) while not rospy.is_shutdown(): try: (trans, rot) = listener.lookupTransform(sys.argv[1], sys.argv[2], rospy.Time(0)) (trans2, rot2) = listener.lookupTransform(sys.argv[3], sys.argv[4], rospy.Time(0)) msg1 = Pose() msg2 = Pose() msg1.position.x, msg1.position.y, msg1.position.z = trans[0], trans[1], trans[2] msg2.position.x, msg2.position.y, msg2.position.z = trans2[0], trans2[1], trans2[2] msg1.orientation.x, msg1.orientation.y, msg1.orientation.z, msg1.orientation.w = rot[0], rot[1], rot[2], rot[3] msg2.orientation.x, msg2.orientation.y, msg2.orientation.z, msg2.orientation.w = rot2[0], rot2[1], rot2[2], rot2[3] (trans_tot, rot_tot) = pm.toTf(pm.fromMsg(msg1)*pm.fromMsg(msg2)) print "translation: ", trans_tot, ", rotation :", rot_tot trans_list.append(trans_tot) rot_list.append(rot_tot) except (tf.LookupException, tf.ConnectivityException): continue rate.sleep() trans_str = str(np.median(np.array(trans_list), axis = 0)) rot_str = str(np.median(np.array(rot_list), axis = 0)) print "median of translation :", trans_str print "median of rotation :", rot_str try: os.remove('../../launch/kinect_playpen_to_torso_lift_link.launch') except: print 'no file to be removed, creating new file' f = open('../../launch/kinect_playpen_to_torso_lift_link.launch', 'w') f.write('<launch>\n') ############################################ I really need to verify the order of sys.arg[4] and sys.arg[1], this could be wrong!!! best way to check ############################################ is to transform something from argv[4] frame to argv[1] frame and check f.write('<node pkg="tf" type="static_transform_publisher" name="kinect_playpen_to_pr2_lift_link" args=" '+trans_str[1:-1]+' '+rot_str[1:-1]+' '+sys.argv[1]+' '+sys.argv[4]+' 30" />\n') f.write('</launch>') f.close() #run this to calibrate playpen, I should include it in a launch file with arguments and launching pr2_launch/ar_kinect and launch/ar_kinect #./calibrate.py /torso_lift_link /calibration_pattern /calibration_pattern2 /openni_camera2
[ [ 1, 0, 0.3563, 0.0115, 0, 0.66, 0, 688, 0, 1, 0, 0, 688, 0, 0 ], [ 1, 0, 0.3678, 0.0115, 0, 0.66, 0.0909, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.3793, 0.0115, 0, ...
[ "import os", "import roslib", "roslib.load_manifest('pr2_playpen')", "roslib.load_manifest('tf_conversions')", "import rospy", "import math", "import tf", "import sys", "import tf_conversions.posemath as pm", "import numpy as np", "from geometry_msgs.msg import Pose", "if __name__ == '__main__...
#!/usr/bin/env python import roslib roslib.load_manifest('pr2_playpen') import rospy import hrl_pr2_lib.pressure_listener as pl import cPickle class PressureWriter: def __init__(self): rospy.init_node('pressure_writer') self.r_grip_press = pl.PressureListener(topic='/pressure/r_gripper_motor') self.l_grip_press = pl.PressureListener(topic='/pressure/l_gripper_motor') self.r_grip_data = [] self.l_grip_data = [] self.status = None def zero(self): self.r_grip_press.rezero() self.l_grip_press.rezero() def record_pressures(self, file_name, arm, time = 10:) file_handle = open(file_name, 'w') self.zero() start = rospy.get_time() #might be better to do some condition, like gripper not moving or when arm is moved to side #while rospy.get_time()-start < time: while not self.status == 'moving somewhere again': print "saving data now ..." self.r_grip_data.append(self.r_grip_press.get_pressure_readings()) self.l_grip_data.append(self.l_grip_press.get_pressure_readings()) rospy.sleep(0.05) cPickle.dump(data, file_handle) file_handle.close() def print_pressures(self): self.r_grip_press.rezero() right_tuple = self.r_grip_press.get_pressure_readings() # print "here's the right gripper :\n", right_tuple # print "here's the raw values : \n", self.r_grip_press.rmat_raw def status_callback(self, data): print "here is the status :", data # self.status = data.feedback.state self.status = data if __name__ == '__main__': pw = PressureWriter() rospy.Subscriber('OverheadServer/actionlib/feedback/feedback/state', String, pw.status_callback) while not rospy.is_shutdown(): if pw.status == 'closing gripper': pw.record_pressures('test_file', 0) rospy.sleep(0.02)
[ [ 1, 0, 0.037, 0.0185, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0556, 0.0185, 0, 0.66, 0.1667, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0741, 0.0185, 0, 0.6...
[ "import roslib", "roslib.load_manifest('pr2_playpen')", "import rospy", "import hrl_pr2_lib.pressure_listener as pl", "import cPickle", "class PressureWriter:\n\n def __init__(self):\n rospy.init_node('pressure_writer')\n self.r_grip_press = pl.PressureListener(topic='/pressure/r_gripper_...
# # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # #Based on some code by Kaijen, modified heavily by Marc Killpack import roslib roslib.load_manifest('pr2_playpen') import rospy from pr2_playpen.pick_and_place_manager import * from object_manipulator.convert_functions import * from pr2_playpen.srv import Playpen from pr2_playpen.srv import Conveyor from pr2_playpen.srv import Check from pr2_playpen.srv import Train from UI_segment_object.srv import Save import os import datetime import cPickle import math import numpy as np class SimplePickAndPlaceExample(): def __init__(self): rospy.loginfo("initializing pick and place manager") self.papm = PickAndPlaceManager() rospy.loginfo("finished initializing pick and place manager") rospy.wait_for_service('playpen') rospy.wait_for_service('conveyor') self.playpen = rospy.ServiceProxy('playpen', Playpen) self.conveyor = rospy.ServiceProxy('conveyor', Conveyor) self.objects_dist = [.135, .26-.135, .405-.26, .545-.405, .70-.545, .865-.70, .995-.865, 1.24-.995] self.tries = 0 self.successes = 0 #pick up the nearest object to PointStamped target_point with whicharm #(0=right, 1=left) def pick_up_object_near_point(self, target_point, whicharm): rospy.loginfo("moving the arms to the side") self.papm.move_arm_out_of_way(0) self.papm.move_arm_out_of_way(1) #############once is it positioned, we don't want to move the head at all !!!############# # rospy.loginfo("pointing the head at the target point") # self.papm.point_head(get_xyz(target_point.point), # target_point.header.frame_id) ######################################################################################### rospy.loginfo("detecting the table and objects") self.papm.call_tabletop_detection(take_static_collision_map = 1, update_table = 1, clear_attached_objects = 1) rospy.loginfo("picking up the nearest object to the target point") success = self.papm.pick_up_object_near_point(target_point, whicharm) if success: rospy.loginfo("pick-up was successful! Moving arm to side") #self.papm.move_arm_to_side(whicharm) self.papm.move_arm_out_of_way(whicharm) else: rospy.loginfo("pick-up failed.") return success #place the object held in whicharm (0=right, 1=left) down in the #place rectangle defined by place_rect_dims (x,y) #and place_rect_center (PoseStamped) def place_object(self, whicharm, place_rect_dims, place_rect_center): self.papm.set_place_area(place_rect_center, place_rect_dims) rospy.loginfo("putting down the object in the %s gripper"\ %self.papm.arm_dict[whicharm]) success = self.papm.put_down_object(whicharm, max_place_tries = 5, use_place_override = 1) if success: rospy.loginfo("place returned success") else: rospy.loginfo("place returned failure") self.papm.open_gripper(whicharm) return success if __name__ == "__main__": rospy.init_node('simple_pick_and_place_example') sppe = SimplePickAndPlaceExample() #adjust for your table table_height = 0.529 date = datetime.datetime.now() f_name = date.strftime("%Y-%m-%d_%H-%M-%S") save_dir = os.getcwd()+'/../../data/'+f_name playpen_dir = '/home/mkillpack/svn/gt-ros-pkg/hrl/pr2_playpen/data/' #should add way to sync os.mkdir(save_dir) print "CHECING FOR DIRECTORY : ", os.getcwd()+'/../../data/'+f_name #.5 m in front of robot, centered target_point_xyz = [.625, 0, table_height] #this is currently an approximation/hack should use ar tag target_point = create_point_stamped(target_point_xyz, 'base_link') arm = 0 rospy.wait_for_service('playpen_train_success') rospy.wait_for_service('playpen_check_success') rospy.wait_for_service('playpen_save_pt_cloud') rospy.wait_for_service('playpen_save_image') # rospy.wait_for_service('pr2_save_pt_cloud') try: train_success = rospy.ServiceProxy('playpen_train_success', Train) check_success = rospy.ServiceProxy('playpen_check_success', Check) # save_pr2_cloud = rospy.ServiceProxy('pr2_save_pt_cloud', Save) save_playpen_cloud = rospy.ServiceProxy('playpen_save_pt_cloud', Save) save_playpen_image = rospy.ServiceProxy('playpen_save_image', Save) except rospy.ServiceException, e: print "Service call failed: %s"%e num_samples = train_success() for i in xrange(len(sppe.objects_dist)): file_handle = open(save_dir+'/object'+str(i).zfill(3)+'.pkl', 'wb') data = {} sppe.playpen(0) sppe.conveyor(sppe.objects_dist[i]) # data['object'+str(i).zfill(3)] = {} # data['object'+str(i).zfill(3)]['success'] = [] # data['object'+str(i).zfill(3)]['frames'] = [] data['success'] = [] data['frames'] = [] while sppe.tries<6: print "arm is ", arm # print "target _point is ", target_point.x # save_pr2_cloud(save_dir+'/object'+str(i).zfill(3)+'_try'+str(sppe.tries).zfill(3)+'_before_pr2.pcd') save_playpen_cloud(playpen_dir+'object'+str(i).zfill(3)+'_try'+str(sppe.tries).zfill(3)+'_before_playpen.pcd') save_playpen_image(playpen_dir+'object'+str(i).zfill(3)+'_try'+str(sppe.tries).zfill(3)+'_before_playpen.png') success = sppe.pick_up_object_near_point(target_point, arm) result = [] rospy.sleep(5) for j in xrange(5): result.append(check_success('').result) rospy.sleep(5) result.sort() if result[2] == 'table': success = True elif result[2] == 'object': success = False # else: # success = False # sppe.tries = sppe.tries-1 # this is to compensate for failures in perception hopefully print "SUCCESS IS :", success, ' ', result data['success'].append(success) # save_pr2_cloud(save_dir+'/object'+str(i).zfill(3)+'try'+str(sppe.tries).zfill(3)+'_after_pr2.pcd') save_playpen_cloud(playpen_dir+'object'+str(i).zfill(3)+'_try'+str(sppe.tries).zfill(3)+'_after_playpen.pcd') save_playpen_image(playpen_dir+'object'+str(i).zfill(3)+'_try'+str(sppe.tries).zfill(3)+'_after_playpen.png') # save_playpen_cloud(save_dir+'/object'+str(i).zfill(3)+'try'+str(sppe.tries).zfill(3)+'_after_playpen.pcd') if success: sppe.successes=sppe.successes + 1 #square of size 30 cm by 30 cm place_rect_dims = [.1, .1] x = np.random.uniform(-0.18, 0.18, 1)[0] y = np.random.uniform(-0.18, 0.18, 1)[0] center_xyz = [.625+x, y, table_height+.10] #aligned with axes of frame_id center_quat = [0,0,0,1] place_rect_center = create_pose_stamped(center_xyz+center_quat, 'base_link') sppe.place_object(arm, place_rect_dims, place_rect_center) arm = arm.__xor__(1) sppe.tries = sppe.tries+1 sppe.playpen(1) sppe.successes = 0 sppe.tries = 0 cPickle.dump(data, file_handle) file_handle.close() # sppe.playpen(0)
[ [ 1, 0, 0.0175, 0.0175, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 1, 0, 0.0351, 0.0175, 0, 0.66, 0.0714, 164, 0, 1, 0, 0, 164, 0, 0 ], [ 1, 0, 0.0526, 0.0175, 0, ...
[ "import roslib", "import rospy", "from pr2_playpen.pick_and_place_manager import *", "from object_manipulator.convert_functions import *", "from pr2_playpen.srv import Playpen", "from pr2_playpen.srv import Conveyor", "from pr2_playpen.srv import Check", "from pr2_playpen.srv import Train", "from UI...
#! /usr/bin/python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # #Based on some code by Kaijen, modified heavily by Marc Killpack import roslib roslib.load_manifest('pr2_playpen') roslib.load_manifest('pr2_grasp_behaviors') import rospy import actionlib import hrl_pr2_lib.pressure_listener as pl from object_manipulator.convert_functions import * from pr2_playpen.srv import Playpen from pr2_playpen.srv import Conveyor from pr2_playpen.srv import Check from pr2_playpen.srv import Train from UI_segment_object.srv import Save #from pr2_overhead_grasping.msg import * from pr2_grasp_behaviors.msg import * import os import datetime import cPickle import math import numpy as np class HRLControllerPlaypen(): def __init__(self): print "waiting for conveyor" rospy.wait_for_service('playpen') rospy.wait_for_service('conveyor') print "started conveyor and playpen" self.playpen = rospy.ServiceProxy('playpen', Playpen) self.conveyor = rospy.ServiceProxy('conveyor', Conveyor) self.objects_dist = [0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13] self.tries = 0 self.successes = 0 self.r_grip_press = pl.PressureListener(topic='/pressure/r_gripper_motor') self.l_grip_press = pl.PressureListener(topic='/pressure/l_gripper_motor') self.grasp_client = [None, None] self.grasp_setup_client = [None, None] self.grasp_client[0] = actionlib.SimpleActionClient('r_overhead_grasp', OverheadGraspAction) self.grasp_client[0].wait_for_server() self.grasp_setup_client[0] = actionlib.SimpleActionClient('r_overhead_grasp_setup', OverheadGraspSetupAction) self.grasp_setup_client[0].wait_for_server() self.grasp_client[1] = actionlib.SimpleActionClient('l_overhead_grasp', OverheadGraspAction) self.grasp_client[1].wait_for_server() self.grasp_setup_client[1] = actionlib.SimpleActionClient('l_overhead_grasp_setup', OverheadGraspSetupAction) self.grasp_setup_client[1].wait_for_server() def move_to_side(self, whicharm, open_gripper=False): rospy.loginfo("moving the arms to the side") setup_goal = OverheadGraspSetupGoal() setup_goal.disable_head = True setup_goal.open_gripper = open_gripper self.grasp_setup_client[whicharm].send_goal(setup_goal) self.grasp_setup_client[whicharm].wait_for_result() #pick up the nearest object to PointStamped target_point with whicharm #(0=right, 1=left) def pick_up_object_near_point(self, target_point, whicharm): self.move_to_side(whicharm, False) #############once is it positioned, we don't want to move the head at all !!!############# # rospy.loginfo("pointing the head at the target point") # self.papm.point_head(get_xyz(target_point.point), # target_point.header.frame_id) ######################################################################################### rospy.loginfo("picking up the nearest object to the target point") ############################################################ # Creating grasp grasp_goal grasp_goal = OverheadGraspGoal() grasp_goal.is_grasp = True grasp_goal.disable_head = True grasp_goal.disable_coll = False grasp_goal.grasp_type=OverheadGraspGoal.VISION_GRASP grasp_goal.grasp_params = [0] * 3 grasp_goal.grasp_params[0] = target_point.point.x grasp_goal.grasp_params[1] = target_point.point.y grasp_goal.behavior_name = "overhead_grasp" grasp_goal.sig_level = 0.999 ############################################################ self.grasp_client[whicharm].send_goal(grasp_goal) self.grasp_client[whicharm].wait_for_result() result = self.grasp_client[whicharm].get_result() success = (result.grasp_result == "Object grasped") if success: rospy.loginfo("pick-up was successful! Moving arm to side") resetup_goal = OverheadGraspSetupGoal() resetup_goal.disable_head = True self.grasp_setup_client[whicharm].send_goal(resetup_goal) self.grasp_setup_client[whicharm].wait_for_result() else: rospy.loginfo("pick-up failed.") return success #place the object held in whicharm (0=right, 1=left) down in the #place rectangle defined by place_rect_dims (x,y) #and place_rect_center (PoseStamped) def place_object(self, whicharm, place_rect_dims, place_rect_center): rospy.loginfo("putting down the object in the r gripper") ############################################################ # Creating place goal grasp_goal = OverheadGraspGoal() grasp_goal.is_grasp = False grasp_goal.disable_head = True grasp_goal.disable_coll = False grasp_goal.grasp_type=OverheadGraspGoal.MANUAL_GRASP grasp_goal.grasp_params = [0] * 3 grasp_goal.grasp_params[0] = place_rect_center.pose.position.x grasp_goal.grasp_params[1] = place_rect_center.pose.position.y grasp_goal.behavior_name = "overhead_grasp" grasp_goal.sig_level = 0.999 ############################################################ self.grasp_client[whicharm].send_goal(grasp_goal) self.grasp_client[whicharm].wait_for_result() result = self.grasp_client[whicharm].get_result() success = (result.grasp_result == "Object placed") if success: rospy.loginfo("place returned success") else: rospy.loginfo("place returned failure") return success if __name__ == "__main__": import optparse p = optparse.OptionParser() p.add_option('--path', action='store', dest='path_save',type='string', default=None, help='this is path to directory for saving files') opt, args = p.parse_args() rospy.init_node('simple_pick_and_place_example') hcp = HRLControllerPlaypen() #adjust for your table table_height = 0.529 date = datetime.datetime.now() f_name = date.strftime("%Y-%m-%d_%H-%M-%S") if opt.path_save == None: print "Not logging or saving data from playpen" SAVE = False else: save_dir = opt.path_save+f_name print "Logging and saving playpen data in :", save_dir SAVE = True if SAVE == True: os.mkdir(save_dir) #print "CHECING FOR DIRECTORY : ", os.getcwd()+'/../../data/'+f_name #.5 m in front of robot, centered target_point_xyz = [.625, 0, table_height] #this is currently an approximation/hack should use ar tag target_point = create_point_stamped(target_point_xyz, 'base_link') arm = 0 rospy.wait_for_service('playpen_train') rospy.wait_for_service('playpen_check_success') if SAVE == True: # rospy.wait_for_service('playpen_save_pt_cloud') # rospy.wait_for_service('playpen_save_image') rospy.wait_for_service('pr2_save_pt_cloud') rospy.wait_for_service('pr2_save_image') try: train = rospy.ServiceProxy('playpen_train', Train) check_success = rospy.ServiceProxy('playpen_check_success', Check) if SAVE == True: save_pr2_cloud = rospy.ServiceProxy('pr2_save_pt_cloud', Save) save_pr2_image = rospy.ServiceProxy('pr2_save_image', Save) # save_playpen_cloud = rospy.ServiceProxy('playpen_save_pt_cloud', Save) # save_playpen_image = rospy.ServiceProxy('playpen_save_image', Save) except rospy.ServiceException, e: print "Service call failed: %s"%e print "moving arms to side" hcp.move_to_side(0, False) hcp.move_to_side(1, False) print "done moving arms, sleeping ..." rospy.sleep(15) print "done sleeping, now training for table top" num_samples = train('table') for i in xrange(len(hcp.objects_dist)): try: if SAVE==True: file_handle = open(save_dir+'/object'+str(i).zfill(3)+'.pkl', 'w') data = {} hcp.playpen(0) hcp.conveyor(hcp.objects_dist[i]) data['success'] = [] data['frames'] = [] data['pressure'] = {} data['pressure']['which_arm'] = [] data['pressure']['data'] = [] start_time = rospy.get_time() # while hcp.tries<3: while rospy.get_time()-start_time < 100.0: print "arm is ", arm hcp.move_to_side(arm, True) rospy.sleep(7) if SAVE == True: save_pr2_cloud(save_dir+'/object'+str(i).zfill(3)+'_try'+str(hcp.tries).zfill(3)+'_before_pr2.pcd') save_pr2_image(save_dir+'/object'+str(i).zfill(3)+'_try'+str(hcp.tries).zfill(3)+'_before_pr2.png') # save_playpen_cloud(playpen_dir+'object'+str(i).zfill(3)+'_try'+str(hcp.tries).zfill(3)+'_before_playpen.pcd') # save_playpen_image(playpen_dir+'object'+str(i).zfill(3)+'_try'+str(hcp.tries).zfill(3)+'_before_playpen.png') num_samples = train('object') print "attempting to pick up the object" success = hcp.pick_up_object_near_point(target_point, arm) print "starting to move arm to side at :", rospy.get_time() hcp.move_to_side(arm, False) print "moved past move to side arm command at :", rospy.get_time() results = [] print "sleeping for 10 seconds, is this necessary ..." rospy.sleep(10) num_samples = 7 for j in xrange(num_samples): results.append(check_success('').result) results.sort() print "results are :", results print "index into result is :", int(num_samples/2) if results[int(num_samples/2)] == 'table': success = True data['success'].append(success) elif results[int(num_samples/2)] == 'object': success = False data['success'].append(success) else: success = None #hcp.tries = hcp.tries-1 # this is to compensate for failures in perception hopefully print "SUCCESS IS :", success if SAVE == True: save_pr2_cloud(save_dir+'/object'+str(i).zfill(3)+'_try'+str(hcp.tries).zfill(3)+'_after_pr2.pcd') save_pr2_image(save_dir+'/object'+str(i).zfill(3)+'_try'+str(hcp.tries).zfill(3)+'_after_pr2.png') # save_playpen_cloud(playpen_dir+'object'+str(i).zfill(3)+'_try'+str(hcp.tries).zfill(3)+'_after_playpen.pcd') # save_playpen_image(playpen_dir+'object'+str(i).zfill(3)+'_try'+str(hcp.tries).zfill(3)+'_after_playpen.png') if success: hcp.successes=hcp.successes + 1 #square of size 30 cm by 30 cm place_rect_dims = [.1, .1] inside = False while inside == False: x = np.random.uniform(-0.18, 0.18, 1)[0] y = np.random.uniform(-0.18, 0.18, 1)[0] if math.sqrt(x*x+y*y) <= 0.18: inside = True center_xyz = [.625+x, y, table_height+.10] #aligned with axes of frame_id center_quat = [0,0,0,1] place_rect_center = create_pose_stamped(center_xyz+center_quat, 'base_link') hcp.place_object(arm, place_rect_dims, place_rect_center) hcp.move_to_side(arm, True) arm = arm.__xor__(1) hcp.tries = hcp.tries+1 hcp.playpen(1) hcp.successes = 0 hcp.tries = 0 cPickle.dump(data, file_handle) file_handle.close() hcp.playpen(0) except: print "failed for object"
[ [ 1, 0, 0.0083, 0.0083, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 1, 0, 0.0167, 0.0083, 0, 0.66, 0.0625, 164, 0, 1, 0, 0, 164, 0, 0 ], [ 1, 0, 0.025, 0.0083, 0, 0...
[ "import roslib", "import rospy", "import actionlib", "import hrl_pr2_lib.pressure_listener as pl", "from object_manipulator.convert_functions import *", "from pr2_playpen.srv import Playpen", "from pr2_playpen.srv import Conveyor", "from pr2_playpen.srv import Check", "from pr2_playpen.srv import Tr...
#!/usr/bin/env python # Calculating and displaying 2D Hue-Saturation histogram of a color image import roslib roslib.load_manifest('opencv2') import sys import cv cv.NamedWindow("back_projection", cv.CV_WINDOW_AUTOSIZE) cv.NamedWindow("back_modified", cv.CV_WINDOW_AUTOSIZE) cv.NamedWindow("back_modified2", cv.CV_WINDOW_AUTOSIZE) def hs_histogram(src, patch): # Convert to HSV hsv = cv.CreateImage(cv.GetSize(src), 8, 3) cv.CvtColor(src, hsv, cv.CV_BGR2HSV) hsv_patch= cv.CreateImage(cv.GetSize(patch), 8, 3) # Extract the H and S planes h_plane = cv.CreateMat(src.rows, src.cols, cv.CV_8UC1) h_plane_img = cv.CreateImage(cv.GetSize(src), 8, 1) h_plane_patch = cv.CreateMat(patch.rows, patch.cols, cv.CV_8UC1) s_plane = cv.CreateMat(src.rows, src.cols, cv.CV_8UC1) s_plane_img = cv.CreateImage(cv.GetSize(src), 8, 1) s_plane_patch = cv.CreateMat(patch.rows, patch.cols, cv.CV_8UC1) v_plane = cv.CreateMat(src.rows, src.cols, cv.CV_8UC1) cv.Split(hsv, h_plane, s_plane, v_plane, None) cv.Split(hsv, h_plane_img, s_plane_img, None, None) cv.Split(hsv_patch, h_plane_patch, s_plane_patch, None, None) #cv.Split(src, h_plane, s_plane, v_plane, None) planes = [h_plane_patch, s_plane_patch]#, s_plane, v_plane] h_bins = 30 s_bins = 32 v_bins = 30 hist_size = [h_bins, s_bins] # hue varies from 0 (~0 deg red) to 180 (~360 deg red again */ h_ranges = [0, 180] # saturation varies from 0 (black-gray-white) to # 255 (pure spectrum color) s_ranges = [0, 255] v_ranges = [0, 255] ranges = [h_ranges, s_ranges]#, s_ranges, v_ranges] scale = 10 hist = cv.CreateHist([h_bins, s_bins], cv.CV_HIST_ARRAY, ranges, 1) cv.CalcHist([cv.GetImage(i) for i in planes], hist) (_, max_value, _, _) = cv.GetMinMaxHistValue(hist) hist_img = cv.CreateImage((h_bins*scale, s_bins*scale), 8, 3) back_proj_img = cv.CreateImage(cv.GetSize(src), 8, 1) cv.CalcBackProject([h_plane_img, s_plane_img], back_proj_img, hist) # for h in range(h_bins): # for s in range(s_bins): # bin_val = cv.QueryHistValue_2D(hist, h, s) # intensity = cv.Round(bin_val * 255 / max_value) # cv.Rectangle(hist_img, # (h*scale, s*scale), # ((h+1)*scale - 1, (s+1)*scale - 1), # cv.RGB(intensity, intensity, intensity), # cv.CV_FILLED) return back_proj_img, hist def back_project_hs(src, patch): # Convert to HSV hsv = cv.CreateImage(cv.GetSize(src), 8, 3) cv.CvtColor(src, hsv, cv.CV_BGR2HSV) hsv_patch= cv.CreateImage(cv.GetSize(patch), 8, 3) cv.CvtColor(patch, hsv_patch, cv.CV_BGR2HSV) # Extract the H and S planes h_plane = cv.CreateMat(src.rows, src.cols, cv.CV_8UC1) h_plane_img = cv.CreateImage(cv.GetSize(src), 8, 1) h_plane_patch = cv.CreateMat(patch.rows, patch.cols, cv.CV_8UC1) s_plane = cv.CreateMat(src.rows, src.cols, cv.CV_8UC1) s_plane_img = cv.CreateImage(cv.GetSize(src), 8, 1) s_plane_patch = cv.CreateMat(patch.rows, patch.cols, cv.CV_8UC1) v_plane = cv.CreateMat(src.rows, src.cols, cv.CV_8UC1) cv.Split(hsv, h_plane, s_plane, v_plane, None) cv.Split(hsv, h_plane_img, s_plane_img, None, None) cv.Split(hsv_patch, h_plane_patch, s_plane_patch, None, None) #cv.Split(src, h_plane, s_plane, v_plane, None) planes = [h_plane_patch, s_plane_patch]#, s_plane, v_plane] # planes = [s_plane_patch]#, s_plane, v_plane] h_bins = 30 s_bins = 32 hist_size = [h_bins, s_bins] # hue varies from 0 (~0 deg red) to 180 (~360 deg red again */ h_ranges = [0, 180] s_ranges = [0, 255] # saturation varies from 0 (black-gray-white) to # 255 (pure spectrum color) ranges = [h_ranges, s_ranges]#, s_ranges, v_ranges] #ranges = [s_ranges]#, s_ranges, v_ranges] scale = 1 hist = cv.CreateHist([h_bins, s_bins], cv.CV_HIST_ARRAY, ranges, 1) #hist = cv.CreateHist([s_bins], cv.CV_HIST_ARRAY, ranges, 1) cv.CalcHist([cv.GetImage(i) for i in planes], hist) (min_value, max_value, _, _) = cv.GetMinMaxHistValue(hist) #cv.NormalizeHist(hist, 20*250.0) print "min hist value is :", min_value print "max hist value is :", max_value back_proj_img = cv.CreateImage(cv.GetSize(src), 8, 1) #cv.NormalizeHist(hist, 2000) cv.CalcBackProject([h_plane_img, s_plane_img], back_proj_img, hist) back_modified = cv.CreateImage(cv.GetSize(src), 8, 1) back_modified2 = cv.CreateImage(cv.GetSize(src), 8, 1) # cv.Dilate(back_proj_img, back_proj_img) # cv.Erode(back_proj_img, back_proj_img) #cv.Smooth(back_proj_img, back_modified) #cv.AdaptiveThreshold(back_proj_img, back_modified, 255, adaptive_method=cv.CV_ADAPTIVE_THRESH_GAUSSIAN_C) #cv.Threshold(back_proj_img, back_modified, 250, 255, cv.CV_THRESH_BINARY) #cv.MorphologyEx(back_modified,back_modified2, None, None, cv.CV_MOP_CLOSE, 3) #cv.MorphologyEx(back_modified,back_modified2, None, None, cv.CV_MOP_CLOSE, 1) # cv.MorphologyEx(back_proj_img,back_modified2, None, None, cv.CV_MOP_CLOSE, 1) #cv.MorphologyEx(back_modified2,back_modified2, None, None, cv.CV_MOP_OPEN, 2) cv.MorphologyEx(back_proj_img,back_modified, None, None, cv.CV_MOP_OPEN, 1) cv.MorphologyEx(back_modified,back_modified, None, None, cv.CV_MOP_CLOSE, 2) cv.Threshold(back_modified, back_modified, 250, 255, cv.CV_THRESH_BINARY) # cv.MorphologyEx(back_proj_img,back_modified2, None, None, cv.CV_MOP_CLOSE, 1) # cv.MorphologyEx(back_modified2,back_modified2, None, None, cv.CV_MOP_OPEN, 2) #cv.FloodFill(back_modified, (320, 240), cv.Scalar(255), cv.Scalar(30), cv.Scalar(30), flags=8) # for i in xrange (10): # cv.MorphologyEx(back_modified,back_modified, None, None, cv.CV_MOP_OPEN, 3) # cv.MorphologyEx(back_modified,back_modified, None, None, cv.CV_MOP_CLOSE, 1) #cv.SubRS(back_modified, 255, back_modified) # cv.CalcBackProject([s_plane_img], back_proj_img, hist) # cv.Scale(back_proj_img, back_proj_img, 30000) cv.ShowImage("back_projection", back_proj_img) cv.ShowImage("back_modified", back_modified) cv.ShowImage("back_modified2", back_modified2) cv.WaitKey(0) #return back_proj_img, hist return back_modified, hist #return , hist if __name__ == '__main__': folder = sys.argv[1] cv.NamedWindow("Source", cv.CV_WINDOW_AUTOSIZE) cv.NamedWindow("final", cv.CV_WINDOW_AUTOSIZE) src2 = cv.LoadImageM(folder+'object'+str(0).zfill(3)+'_try'+str(0).zfill(3)+'_after_pr2.png') patch_images = [] avg_noise = cv.CreateImage(cv.GetSize(src2), 8, 1) cv.Zero(avg_noise) for k in xrange(1): patch_images.append(cv.LoadImageM('/home/mkillpack/Desktop/patch2.png')) #for i in [4]: for i in xrange(9): for j in xrange(100): print folder+'object'+str(i).zfill(3)+'_try'+str(j).zfill(3)+'_after_pr2.png' src = cv.LoadImageM(folder+'object'+str(i).zfill(3)+'_try'+str(j).zfill(3)+'_after_pr2.png') cv.ShowImage("Source", src) back_proj_img, hist1 = back_project_hs(src, patch_images[0]) back_proj_img2, hist2 = back_project_hs(src2, patch_images[0]) scratch = cv.CreateImage(cv.GetSize(back_proj_img2), 8, 1) scratch2 = cv.CreateImage(cv.GetSize(back_proj_img2), 8, 1) # do something clever with ands ors and diffs cv.Zero(scratch) cv.Zero(scratch2) #idea is to have a background model from back_proj_img2, or at least an emtpy single shot ###cv.Sub(back_proj_img, back_proj_img2, scratch) #cv.SubRS(back_proj_img, 255, scratch) ###cv.SubRS(back_proj_img2, 255, scratch2) #cv.Sub(back_proj_img, back_proj_img2, scratch2) #opposite noise, but excludes object cv.Sub(back_proj_img2, back_proj_img, scratch2) #noise, but includes object if failed, #would need to learn before then update selectively #Maybe want both added in the end. cv.Sub(scratch2, avg_noise, scratch) cv.Or(avg_noise, scratch2, avg_noise) ##adding this part fills in wherever the object has been too, heatmaps? #cv.Sub(back_proj_img2, back_proj_img, scratch) #cv.Or(avg_noise, scratch, avg_noise) # #cv.Sub(back_proj_img2, avg_noise, back_proj_img2) #cv.Sub(scratch,, back_proj_img2) cv.ShowImage("final", scratch) #cv.Sub(scratch, avg_noise, scratch2) #cv.And(scratch, back_proj_img2, scratch2) #cv.SubRS(scratch2, 255, scratch) #cv.ShowImage("final", back_proj_img) print cv.CompareHist(hist1, hist2, cv.CV_COMP_BHATTACHARYYA) #making a mask #mask = cv.CreateImage(cv.GetSize(back_proj_img2), 8, 1) #cv.SubRS(back_proj_img2, 255, back_proj_img2) #cv.SubRS(back_proj_img, 255, back_proj_img, mask=back_proj_img2) #cv.SubRS(back_proj_img, 255, back_proj_img) #cv.MorphologyEx(back_proj_img,back_proj_img, None, None, cv.CV_MOP_OPEN, 8) #cv.MorphologyEx(back_proj_img,back_proj_img, None, None, cv.CV_MOP_CLOSE, 8) #cv.ShowImage("back_projection", back_proj_img2) #cv.WaitKey(0) cv.Scale(back_proj_img, back_proj_img, 1/255.0) print "here's the sum :", cv.Sum(scratch2)
[ [ 1, 0, 0.0126, 0.0042, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0167, 0.0042, 0, 0.66, 0.1111, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0209, 0.0042, 0, 0....
[ "import roslib", "roslib.load_manifest('opencv2')", "import sys", "import cv", "cv.NamedWindow(\"back_projection\", cv.CV_WINDOW_AUTOSIZE)", "cv.NamedWindow(\"back_modified\", cv.CV_WINDOW_AUTOSIZE)", "cv.NamedWindow(\"back_modified2\", cv.CV_WINDOW_AUTOSIZE)", "def hs_histogram(src, patch):\n # Co...
#!/usr/bin/env python # Calculating and displaying 2D Hue-Saturation histogram of a color image import roslib roslib.load_manifest('opencv2') import sys import cv import numpy as np from collections import deque cv.NamedWindow("avg_noise", cv.CV_WINDOW_AUTOSIZE) cv.NamedWindow("back_modified", cv.CV_WINDOW_AUTOSIZE) cv.NamedWindow("back_modified2", cv.CV_WINDOW_AUTOSIZE) class HistAnalyzer: def __init__(self, background_noise, mask): self.background_noise = background_noise self.h_bins = 30 self.s_bins = 32 self.h_ranges = [0, 180] self.s_ranges = [0, 255] self.ranges = [self.h_ranges, self.s_ranges] self.hist = None self.mask = mask self.avg_noise = None def calc_hist(self): self.hist = cv.CreateHist([self.h_bins, self.s_bins], cv.CV_HIST_ARRAY, self.ranges, 1) hsv = cv.CreateImage(cv.GetSize(self.background_noise[0]), 8, 3) h_plane = cv.CreateMat(self.background_noise[0].height, self.background_noise[0].width, cv.CV_8UC1) s_plane = cv.CreateMat(self.background_noise[0].height, self.background_noise[0].width, cv.CV_8UC1) for i in xrange(len(self.background_noise)): cv.CvtColor(self.background_noise[i], hsv, cv.CV_BGR2HSV) cv.Split(hsv, h_plane, s_plane, None, None) planes = [h_plane, s_plane]#, s_plane, v_plane] cv.CalcHist([cv.GetImage(i) for i in planes], self.hist, True, self.mask) #cv.NormalizeHist(self.hist, 1.0) def check_for_hist(self): if self.hist == None: print "YOU CAN'T CALCULATE NOISE WITH HIST MODEL OF TABLETOP" exit def calc_noise(self): self.check_for_hist() self.avg_noise = cv.CreateImage(cv.GetSize(self.background_noise[0]), 8, 1) cv.Zero(self.avg_noise) for i in xrange(len(self.background_noise)-1): back_proj_img1, hist1 = self.back_project_hs(self.background_noise[i]) back_proj_img2, hist2 = self.back_project_hs(self.background_noise[i+1]) scratch = cv.CreateImage(cv.GetSize(back_proj_img2), 8, 1) scratch2 = cv.CreateImage(cv.GetSize(back_proj_img2), 8, 1) # do something clever with ands ors and diffs cv.Zero(scratch) cv.Zero(scratch2) cv.Sub(back_proj_img2, back_proj_img1, scratch2) #noise, but includes object if failed, cv.Sub(scratch2, self.avg_noise, scratch) cv.Or(self.avg_noise, scratch2, self.avg_noise) cv.WaitKey(100) def compare_imgs(self, img1, img2): back_proj_img, hist1 = self.back_project_hs(img1) back_proj_img2, hist2 = self.back_project_hs(img2) scratch = cv.CreateImage(cv.GetSize(back_proj_img2), 8, 1) scratch2 = cv.CreateImage(cv.GetSize(back_proj_img2), 8, 1) cv.Zero(scratch) cv.Zero(scratch2) #cv.Sub(back_proj_img, back_proj_img2, scratch2) #opposite noise, but excludes object cv.Sub(back_proj_img2, back_proj_img, scratch2) #noise, but includes object if failed, cv.Sub(scratch2, ha.avg_noise, scratch) return scratch def back_project_hs(self, img): self.check_for_hist() hsv = cv.CreateImage(cv.GetSize(img), 8, 3) scratch = cv.CreateImage(cv.GetSize(img), 8, 1) back_proj_img = cv.CreateImage(cv.GetSize(img), 8, 1) cv.CvtColor(img, hsv, cv.CV_BGR2HSV) h_plane_img = cv.CreateImage(cv.GetSize(img), 8, 1) s_plane_img = cv.CreateImage(cv.GetSize(img), 8, 1) cv.Split(hsv, h_plane_img, s_plane_img, None, None) cv.CalcBackProject([h_plane_img, s_plane_img], back_proj_img, self.hist) cv.MorphologyEx(back_proj_img, back_proj_img, None, None, cv.CV_MOP_OPEN, 1) cv.MorphologyEx(back_proj_img, back_proj_img, None, None, cv.CV_MOP_CLOSE, 2) cv.Threshold(back_proj_img, back_proj_img, 250, 255, cv.CV_THRESH_BINARY) return back_proj_img, self.hist if __name__ == '__main__': folder = sys.argv[1]+'/background_noise/' background_noise = deque() #[] cv.NamedWindow("Source", cv.CV_WINDOW_AUTOSIZE) cv.NamedWindow("final", cv.CV_WINDOW_AUTOSIZE) for i in xrange(130): background_noise.append(cv.LoadImage(folder+'file'+str(i).zfill(3)+'.png')) mask = cv.LoadImage(sys.argv[2], 0) ha = HistAnalyzer(background_noise, mask) ha.calc_hist() ha.calc_noise() back_sum_ls = deque() #[] for i in xrange(130): img = cv.LoadImage(folder+'file'+str(i).zfill(3)+'.png') result = ha.compare_imgs(img, ha.background_noise[0]) back_sum_ls.append(float(cv.Sum(result)[0])) avg = np.mean(back_sum_ls) std = np.std(back_sum_ls) print "avg and std are :", avg, std n = 0 sum_val = 0 #test_sum_ls = [] for i in xrange(9): for j in xrange(100): #print sys.argv[1]+'/object'+str(i).zfill(3)+'_try'+str(j).zfill(3) try: img = cv.LoadImageM(sys.argv[1]+'/object'+str(i).zfill(3)+'_try'+str(j).zfill(3)+'_after_pr2.png') #img = cv.LoadImageM(sys.argv[1]+'/object'+str(i).zfill(3)+'_try'+str(j).zfill(3)+'_before_pr2.png') result = ha.compare_imgs(img, ha.background_noise[-1]) n = n+1 sum_val = sum_val + float(cv.Sum(result)[0]) #print "here's the sum :", cv.Sum(result)[0] cv.ShowImage("Source", img) cv.ShowImage("final", result) cv.WaitKey(-1) loc_sum = float(cv.Sum(result)[0]) #if loc_sum > avg-5*std and loc_sum < avg+5*std: if loc_sum < avg+5*std: print "success ! \t:", loc_sum, "\t compared to \t", avg, 5*std #ha.background_noise.popleft() #ha.background_noise.append(img) else: print "epic fail ! \t:", loc_sum, "\t compared to \t", avg, 5*std except: print "no file like that, probably outside of index range" print "recalculating hist and noise..." #ha.calc_hist() #ha.calc_noise() print "done!" print "average error for objects present :", sum_val/n
[ [ 1, 0, 0.0186, 0.0062, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0248, 0.0062, 0, 0.66, 0.1, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0311, 0.0062, 0, 0.66,...
[ "import roslib", "roslib.load_manifest('opencv2')", "import sys", "import cv", "import numpy as np", "from collections import deque", "cv.NamedWindow(\"avg_noise\", cv.CV_WINDOW_AUTOSIZE)", "cv.NamedWindow(\"back_modified\", cv.CV_WINDOW_AUTOSIZE)", "cv.NamedWindow(\"back_modified2\", cv.CV_WINDOW_A...
#!/usr/bin/env python import roslib roslib.load_manifest('pr2_playpen') import rospy import math import tf import numpy as np import os import sys import cPickle as pkl def is_topic_pub(topic): flag = False for item in rospy.get_published_topics(): for thing in item: if topic in thing: flag = True else: pass return flag def get_data(listener, rate): right = False left = False dist_ls = [] time_ls = [] while is_topic_pub('/r_overhead_grasp/feedback') == False and is_topic_pub('/l_overhead_grasp/feedback') == False: print "waiting for bag file" rate.sleep() if is_topic_pub('/r_overhead_grasp/feedback'): right = True elif is_topic_pub('/l_overhead_grasp/feedback'): left = True if left == True: prefix = 'l_' elif right == True: prefix = 'r_' frame1 = prefix+'gripper_l_finger_tip_link' frame2 = prefix+'gripper_r_finger_tip_link' #run = True while is_topic_pub('/clock') == True: #run == True:#not rospy.is_shutdown(): try: (trans,rot) = listener.lookupTransform(frame1, frame2, rospy.Time(0)) dist = math.sqrt((np.matrix(trans)*np.matrix(trans).T)[0,0]) time = rospy.get_time() dist_ls.append(dist) time_ls.append(time) except (tf.LookupException, tf.ConnectivityException): #run = False continue rate.sleep() return dist_ls, time_ls if __name__ == '__main__': rospy.init_node('get_gripper_position') listener = tf.TransformListener() rate = rospy.Rate(100.0) path = sys.argv[1] print "path is :", path for i in xrange(9): j = 0 dist_dict = {} f_hand = open(path+'/object'+str(i).zfill(3)+'_gripper_dist.pkl', 'w') while os.path.isfile(path + '/object'+str(i).zfill(3)+'_try'+str(j).zfill(3)+'.bag') == True: #j < 999: f_path = path + '/object'+str(i).zfill(3)+'_try'+str(j).zfill(3)+'.bag' os.system('rosbag play -r 2 '+ f_path + ' &') dist_ls, time_ls = get_data(listener, rate) dist_dict['try'+str(j).zfill(3)] = {'dist':dist_ls, 'time':time_ls} j = j+1 pkl.dump(dist_dict, f_hand) f_hand.close()
[ [ 1, 0, 0.0241, 0.012, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0361, 0.012, 0, 0.66, 0.0909, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0482, 0.012, 0, 0.66,...
[ "import roslib", "roslib.load_manifest('pr2_playpen')", "import rospy", "import math", "import tf", "import numpy as np", "import os", "import sys", "import cPickle as pkl", "def is_topic_pub(topic):\n flag = False\n for item in rospy.get_published_topics():\n for thing in item:\n ...
#!/usr/bin/env python import roslib roslib.load_manifest('pr2_playpen') import rospy from sensor_msgs.msg import PointCloud2 import point_cloud_python as pc2py import numpy as np from matplotlib import pylab as pl import draw_scene as ds import math from pr2_playpen.srv import * #this is for Train and Check import threading class ResultsAnalyzer: def __init__(self): rospy.init_node('playpen_results') self.draw = ds.SceneDraw() self.cloud = None rospy.Subscriber("pr2_segment_region", PointCloud2, self.callback) # rospy.Subscriber("playpen_segment_object", PointCloud2, self.callback) self.check = rospy.Service("playpen_check_success", Check, self.serv_success) self.train_empty = rospy.Service("playpen_train", Train, self.serv_train) self.table_mean = None self.table_cov = None self.object_mean = None self.object_cov = None self.mean_ls =[] self.cov_ls = [] self.start_time = rospy.get_time() self.lock = threading.RLock() self.new_cloud = False self.start = 0 def callback(self, data): self.lock.acquire() self.cloud = list(pc2py.points(data, ['x', 'y', 'z'])) self.new_cloud = True print rospy.get_time()-self.start, "time in between call backs" self.start = rospy.get_time() self.lock.release() # self.get_cloud_stats() def serv_success(self, req): result = "none" while self.new_cloud == False: rospy.sleep(0.01) self.new_cloud = False self.lock.acquire() if self.table_mean == None or self.object_mean == None: print "you haven't initialized yet!!" return CheckResponse(result) np_array_cloud = np.array(self.cloud) f_ind = np.array(~np.isnan(np_array_cloud).any(1)).flatten() f_np_array_cloud = np_array_cloud[f_ind, :] if np.max(f_np_array_cloud.shape)>200: mean_3d = np.mean(f_np_array_cloud, axis = 0) cov_3d = np.cov(f_np_array_cloud.T) v, d = np.linalg.eig(cov_3d) max_var = d[:, v == np.max(v)] mean_dist_table = (np.matrix(mean_3d).reshape(3,1)-self.table_mean) mean_dist_object = (np.matrix(mean_3d).reshape(3,1)-self.object_mean) mahal_dist_table = mean_dist_table.T*0.5*np.matrix(np.linalg.inv(cov_3d)+np.linalg.inv(self.table_cov))*mean_dist_table mahal_dist_object = mean_dist_object.T*0.5*np.matrix(np.linalg.inv(cov_3d)+np.linalg.inv(self.object_cov))*mean_dist_object print "table distance = ", mahal_dist_table print "object distance = ", mahal_dist_object # print "d = ", d # print "v = ", v #made a simple change so that I only look if table is empty or not #comparison is made from the user selected region with a base model #of mean covariance of the empty tabletop if np.max(f_np_array_cloud.shape)<200: result = "no_cloud" elif mahal_dist_table<mahal_dist_object: result = "table" else: result = "object" # if req.exp_state == 'empty': # if np.max(f_np_array_cloud.shape)<200: # result = "success" # elif mahal_dist<5*self.nom_dist: # result = "success" # else: # result = "fail" # elif req.exp_state == 'object': # if np.max(f_np_array_cloud.shape)<200: # result = "fail" # elif mahal_dist<5*self.nom_dist: # result = "success" # else: # result = "fail" # elif req.exp_state == 'objects': # print "multiple objects is not yet supported" self.lock.release() return CheckResponse(result) def serv_train(self, req): num_samples = 0 if req.expected == 'table': self.table_mean = None self.table_cov = None print "training for empty table top" elif req.expected == 'object': self.object_mean = None self.object_cov = None print "training for object on table top" self.mean_ls = [] self.cov_ls = [] while num_samples < 11: start_time = rospy.get_time() while self.new_cloud == False: rospy.sleep(0.01) self.lock.acquire() np_array_cloud = np.array(self.cloud) f_ind = np.array(~np.isnan(np_array_cloud).any(1)).flatten() f_np_array_cloud = np_array_cloud[f_ind, :] if np.max(f_np_array_cloud.shape)>200: mean_3d = np.mean(f_np_array_cloud, axis = 0) cov_3d = np.cov(f_np_array_cloud.T) v, d = np.linalg.eig(cov_3d) max_var = d[:, v == np.max(v)] self.mean_ls.append(np.matrix(mean_3d).reshape(3,1)) self.cov_ls.append(np.matrix(cov_3d)) num_samples = num_samples + 1 self.new_cloud = False print "still initializing" self.lock.release() buf_mean = np.matrix(np.zeros((3,1))) buf_cov = np.matrix(np.zeros((3,3))) print "here is the mean list :", self.mean_ls mean_arr = np.array(self.mean_ls) mean_arr.sort(axis=0) print "here is th sorted mean array :", mean_arr print "here is the mean cov :\n", self.cov_ls cov_arr = np.array(self.cov_ls) cov_arr.sort(axis=0) print "here is the sorted cov :\n", cov_arr # for i in xrange(10): # buf_mean = buf_mean + self.mean_ls[i] # buf_cov = buf_cov + self.cov_ls[i] #this is not exactly correct if populations #have different # of points, but it should be approximately right if req.expected == 'table': self.table_mean = mean_arr[5] self.table_cov = cov_arr[5] # self.table_mean = buf_mean*1/10.0 # self.table_cov = buf_cov*1/10.0 elif req.expected == 'object': self.object_mean = mean_arr[5] self.object_cov = cov_arr[5] # self.object_mean = buf_mean*1/10.0 # self.object_cov = buf_cov*1/10.0 return TrainResponse(num_samples) def run(self): print 'Ready to calculate results' rospy.spin() def get_cloud_stats(self): np_array_cloud = np.array(self.cloud) f_ind = np.array(~np.isnan(np_array_cloud).any(1)).flatten() f_np_array_cloud = np_array_cloud[f_ind, :] print 'size of remaining point cloud :', np.max(f_np_array_cloud.shape) if np.max(f_np_array_cloud.shape)>200: mean_3d = np.mean(f_np_array_cloud, axis = 0) cov_3d = np.cov(f_np_array_cloud.T) v, d = np.linalg.eig(cov_3d) max_var = d[:, v == np.max(v)] mean_dist = (np.matrix(mean_3d).reshape(3,1)-self.nom_mean) mahal_dist = mean_dist.T*0.5*np.matrix(np.linalg.inv(cov_3d)+np.linalg.inv(self.nom_cov))*mean_dist print "distance = ", mahal_dist if rospy.get_time()-self.start_time < 10: self.nom_mean = np.matrix(mean_3d).reshape(3,1) self.nom_cov = np.matrix(cov_3d) print "still initializing" else: print "real distance now" else: print "not doing anything since point cloud is too small" # print "mean :\n", mean_3d # print "cov matrix :\n", cov_3d # print "eig of cov :\n", v # print d # print "max direction of var :\n", max_var # hs.draw.pub_body((0, 0, 0), (0, 0, 0, 1), # (0.2, 0.2, 0.2), (1, 0,0,1), 1000000, hs.draw.Marker.SPHERE) # self.draw.pub_body((mean_3d[2], -1*mean_3d[0], -1*mean_3d[1]), (0, 0, 0, 1), # (0.2, 0.2, 0.2), (0, 1,0,1), 1000001, self.draw.Marker.SPHERE) if __name__ == '__main__': ra = ResultsAnalyzer() ra.run() # import roslib # roslib.load_manifest('my_package') # import sys # import rospy # import cv # from std_msgs.msg import String # from sensor_msgs.msg import Image # from cv_bridge import CvBridge, CvBridgeError # class image_converter: # def __init__(self): # self.image_pub = rospy.Publisher("image_topic_2",Image) # cv.NamedWindow("Image window", 1) # self.bridge = CvBridge() # self.image_sub = rospy.Subscriber("image_topic",Image,self.callback) # def callback(self,data): # try: # cv_image = self.bridge.imgmsg_to_cv(data, "bgr8") # except CvBridgeError, e: # print e # (cols,rows) = cv.GetSize(cv_image) # if cols > 60 and rows > 60 : # cv.Circle(cv_image, (50,50), 10, 255) # cv.ShowImage("Image window", cv_image) # cv.WaitKey(3) # try: # self.image_pub.publish(self.bridge.cv_to_imgmsg(cv_image, "bgr8")) # except CvBridgeError, e: # print e # def main(args): # ic = image_converter() # rospy.init_node('image_converter', anonymous=True) # try: # rospy.spin() # except KeyboardInterrupt: # print "Shutting down" # cv.DestroyAllWindows() # if __name__ == '__main__': # main(sys.argv)
[ [ 1, 0, 0.0076, 0.0038, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0115, 0.0038, 0, 0.66, 0.0833, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0153, 0.0038, 0, 0....
[ "import roslib", "roslib.load_manifest('pr2_playpen')", "import rospy", "from sensor_msgs.msg import PointCloud2", "import point_cloud_python as pc2py", "import numpy as np", "from matplotlib import pylab as pl", "import draw_scene as ds", "import math", "from pr2_playpen.srv import * #this is for...
import roslib roslib.load_manifest('opencv2') import cv a = cv.LoadImage('/home/mkillpack/hrl_file_server/playpen_data_sets/2011-06-30_19-01-02/object000_try011_before_pr2.png', 0) b = cv.LoadImage('/home/mkillpack/hrl_file_server/playpen_data_sets/2011-06-30_19-01-02/object000_try011_after_pr2.png', 0) foreground = cv.CreateImage((640,480), 8, 1) size = cv.GetSize(a) IavgF = cv.CreateImage(size, cv.IPL_DEPTH_32F, 3) IdiffF = cv.CreateImage(size, cv.IPL_DEPTH_32F, 3) IprevF = cv.CreateImage(size, cv.IPL_DEPTH_32F, 3) IhiF = cv.CreateImage(size, cv.IPL_DEPTH_32F, 3) IlowF = cv.CreateImage(size, cv.IPL_DEPTH_32F, 3) Ilow1 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 1) Ilow2 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 1) Ilow3 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 1) Ihi1 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 1) Ihi2 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 1) Ihi3 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 1) cv.Zero(IavgF) cv.Zero(IdiffF) cv.Zero(IprevF) cv.Zero(IhiF) cv.Zero(IlowF) Icount = 0.00001 Iscratch = cv.CreateImage(size, cv.IPL_DEPTH_32F, 3) Iscratch2 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 3) Igray1 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 1) Igray2 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 1) Igray3 = cv.CreateImage(size, cv.IPL_DEPTH_32F, 1) Imaskt = cv.CreateImage(size, cv.IPL_DEPTH_8U, 1) cv.Zero(Iscratch) cv.Zero(Iscratch2) first = 1 def accumulateBackground(img): global first, Icount cv.CvtScale(img, Iscratch, 1, 0) if (not first): cv.Acc(Iscratch, IavgF) cv.AbsDiff(Iscratch, IprevF, Iscratch2) cv.Acc(Iscratch2, IdiffF) Icount += 1.0 first = 0 cv.Copy(Iscratch, IprevF) def setHighThresh(thresh): cv.ConvertScale(IdiffF, Iscratch, thresh) cv.Add(Iscratch, IavgF, IhiF) cv.Split(IhiF, Ihi1, Ihi2, Ihi3, None) def setLowThresh(thresh): cv.ConvertScale(IdiffF, Iscratch, thresh) cv.Sub(IavgF, Iscratch, IlowF) cv.Split(IlowF, Ilow1, Ilow2, Ilow3, None) def createModelsfromStats(): cv.ConvertScale(IavgF, IavgF, float(1.0/Icount)) cv.ConvertScale(IdiffF, IdiffF, float(1.0/Icount)) cv.AddS(IdiffF, cv.Scalar(1.0, 1.0, 1.0), IdiffF) setHighThresh(10.0) setLowThresh(10.0) def backgroundDiff(img, Imask): cv.CvtScale(img, Iscratch, 1, 0) cv.Split(Iscratch, Igray1, Igray2, Igray3, None) cv.InRange(Igray1, Ilow1, Ihi1, Imask) cv.InRange(Igray2, Ilow2, Ihi2, Imaskt) cv.Or(Imask, Imaskt, Imask) cv.InRange(Igray3, Ilow3, Ihi3, Imaskt) cv.Or(Imask, Imaskt, Imask) cv.SubRS(Imask, 255, Imask) cv.SaveImage('/home/mkillpack/Desktop/mask.png', Imask) #cv.Erode(Imask, Imask) print "here is the sum of the non-zero pixels", cv.Sum(Imask) return Imask if __name__ == '__main__': folder = '/home/mkillpack/hrl_file_server/playpen_data_sets/2011-06-30_19-01-02/' for j in [3]: #[0, 3, 4, 5, 6] for i in xrange(200): try: file_name = folder+'object'+str(j).zfill(3)+'_try'+str(i).zfill(3)+'_after_pr2.png' img = cv.LoadImage(file_name, 1) print "reading ", file_name, '...' except: print file_name, " doesn't exist" if not img == None: accumulateBackground(img) #c = cv.LoadImage('/home/mkillpack/hrl_file_server/playpen_data_sets/2011-06-30_19-01-02/object000_try012_after_pr2.png', 1) createModelsfromStats() file_name = folder+'object006_try'+str(0).zfill(3)+'_before_pr2.png' img = cv.LoadImage(file_name, 3) Imask = cv.CreateImage(size, cv.IPL_DEPTH_8U, 1) cv.Zero(Imask) cv.Zero(Imaskt) Imask = backgroundDiff(img, Imask)
[ [ 1, 0, 0.0093, 0.0093, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.0185, 0.0093, 0, 0.66, 0.0263, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.0278, 0.0093, 0, 0....
[ "import roslib", "roslib.load_manifest('opencv2')", "import cv", "a = cv.LoadImage('/home/mkillpack/hrl_file_server/playpen_data_sets/2011-06-30_19-01-02/object000_try011_before_pr2.png', 0)", "b = cv.LoadImage('/home/mkillpack/hrl_file_server/playpen_data_sets/2011-06-30_19-01-02/object000_try011_after_pr2...
####################################################################### # # USE pr2_object_manipulation/pr2_gripper_reactive_approach/controller_manager.py # That code has much of the ideas at the bottom, with more. # ####################################################################### # TODO Update code to throw points one at a time. Sections are labled: "Hack" import numpy as np, math from threading import RLock, Timer import sys import roslib; roslib.load_manifest('hrl_pr2_lib') import tf import rospy import actionlib from actionlib_msgs.msg import GoalStatus from pr2_controllers_msgs.msg import Pr2GripperCommandGoal, Pr2GripperCommandAction, Pr2GripperCommand from geometry_msgs.msg import PoseStamped from teleop_controllers.msg import JTTeleopControllerState from std_msgs.msg import Float64 from sensor_msgs.msg import JointState import hrl_lib.transforms as tr import time import tf.transformations as tftrans import types node_name = "pr2_arms" def log(str): rospy.loginfo(node_name + ": " + str) ## # Class for simple management of the arms and grippers. # Provides functionality for moving the arms, opening and closing # the grippers, performing IK, and other functionality as it is # developed. class PR2Arms(object): ## # Initializes all of the servers, clients, and variables # # @param send_delay send trajectory points send_delay nanoseconds into the future # @param gripper_point given the frame of the wrist_roll_link, this point offsets # the location used in FK and IK, preferably to the tip of the # gripper def __init__(self, send_delay=50000000, gripper_point=(0.23,0.0,0.0), force_torque = False): log("Loading PR2Arms") self.send_delay = send_delay self.off_point = gripper_point self.gripper_action_client = [actionlib.SimpleActionClient('r_gripper_controller/gripper_action', Pr2GripperCommandAction),actionlib.SimpleActionClient('l_gripper_controller/gripper_action', Pr2GripperCommandAction)] self.gripper_action_client[0].wait_for_server() self.gripper_action_client[1].wait_for_server() self.arm_state_lock = [RLock(), RLock()] self.r_arm_cart_pub = rospy.Publisher('/r_cart/command_pose', PoseStamped) self.l_arm_cart_pub = rospy.Publisher('/l_cart/command_pose', PoseStamped) rospy.Subscriber('/r_cart/state', JTTeleopControllerState, self.r_cart_state_cb) rospy.Subscriber('/l_cart/state', JTTeleopControllerState, self.l_cart_state_cb) self.tf_lstnr = tf.TransformListener() rospy.sleep(1.) log("Finished loading SimpleArmManger") def r_cart_state_cb(self, msg): trans, quat = self.tf_lstnr.lookupTransform('/torso_lift_link', 'r_gripper_tool_frame', rospy.Time(0)) rot = tr.quaternion_to_matrix(quat) tip = np.matrix([0.12, 0., 0.]).T self.r_ee_pos = rot*tip + np.matrix(trans).T self.r_ee_rot = rot ros_pt = msg.x_desi_filtered.pose.position x, y, z = ros_pt.x, ros_pt.y, ros_pt.z self.r_cep_pos = np.matrix([x, y, z]).T pt = rot.T * (np.matrix([x,y,z]).T - np.matrix(trans).T) pt = pt + tip self.r_cep_pos_hooktip = rot*pt + np.matrix(trans).T ros_quat = msg.x_desi_filtered.pose.orientation quat = (ros_quat.x, ros_quat.y, ros_quat.z, ros_quat.w) self.r_cep_rot = tr.quaternion_to_matrix(quat) def l_cart_state_cb(self, msg): ros_pt = msg.x_desi_filtered.pose.position self.l_cep_pos = np.matrix([ros_pt.x, ros_pt.y, ros_pt.z]).T ros_quat = msg.x_desi_filtered.pose.orientation quat = (ros_quat.x, ros_quat.y, ros_quat.z, ros_quat.w) self.l_cep_rot = tr.quaternion_to_matrix(quat) def get_ee_jtt(self, arm): if arm == 0: return self.r_ee_pos, self.r_ee_rot else: return self.l_ee_pos, self.l_ee_rot def get_cep_jtt(self, arm, hook_tip = False): if arm == 0: if hook_tip: return self.r_cep_pos_hooktip, self.r_cep_rot else: return self.r_cep_pos, self.r_cep_rot else: return self.l_cep_pos, self.l_cep_rot # set a cep using the Jacobian Transpose controller. def set_cep_jtt(self, arm, p, rot=None): if arm != 1: arm = 0 ps = PoseStamped() ps.header.stamp = rospy.rostime.get_rostime() ps.header.frame_id = 'torso_lift_link' ps.pose.position.x = p[0,0] ps.pose.position.y = p[1,0] ps.pose.position.z = p[2,0] if rot == None: if arm == 0: rot = self.r_cep_rot else: rot = self.l_cep_rot quat = tr.matrix_to_quaternion(rot) ps.pose.orientation.x = quat[0] ps.pose.orientation.y = quat[1] ps.pose.orientation.z = quat[2] ps.pose.orientation.w = quat[3] if arm == 0: self.r_arm_cart_pub.publish(ps) else: self.l_arm_cart_pub.publish(ps) # rotational interpolation unimplemented. def go_cep_jtt(self, arm, p): step_size = 0.01 sleep_time = 0.1 cep_p, cep_rot = self.get_cep_jtt(arm) unit_vec = (p-cep_p) unit_vec = unit_vec / np.linalg.norm(unit_vec) while np.linalg.norm(p-cep_p) > step_size: cep_p += unit_vec * step_size self.set_cep_jtt(arm, cep_p) rospy.sleep(sleep_time) self.set_cep_jtt(arm, p) rospy.sleep(sleep_time) # TODO Evaluate gripper functions and parameters ## # Move the gripper the given amount with given amount of effort # # @param arm 0 for right, 1 for left # @param amount the amount the gripper should be opened # @param effort - supposed to be in Newtons. (-ve number => max effort) def move_gripper(self, arm, amount=0.08, effort = 15): self.gripper_action_client[arm].send_goal(Pr2GripperCommandGoal(Pr2GripperCommand(position=amount, max_effort = effort))) ## # Open the gripper # # @param arm 0 for right, 1 for left def open_gripper(self, arm): self.move_gripper(arm, 0.08, -1) ## # Close the gripper # # @param arm 0 for right, 1 for left def close_gripper(self, arm, effort = 15): self.move_gripper(arm, 0.0, effort) # def get_wrist_force(self, arm): # pass ###################################################### # More specific functionality ###################################################### if __name__ == '__main__': rospy.init_node(node_name, anonymous = True) log("Node initialized") pr2_arm = PR2Arms() # #------- testing set JEP --------------- # raw_input('Hit ENTER to begin') r_arm, l_arm = 0, 1 # cep_p, cep_rot = pr2_arm.get_cep_jtt(r_arm) # print 'cep_p:', cep_p.A1 # # for i in range(5): # cep_p[0,0] += 0.01 # raw_input('Hit ENTER to move') # pr2_arm.set_cep_jtt(r_arm, cep_p) raw_input('Hit ENTER to move') p1 = np.matrix([0.62, 0.0, 0.16]).T pr2_arm.go_cep_jtt(r_arm, p1) #rospy.sleep(10) #pr2_arm.close_gripper(r_arm, effort = -1) raw_input('Hit ENTER to move') p2 = np.matrix([0.600+0.06, 0.106, -0.32]).T pr2_arm.go_cep_jtt(r_arm, p2) raw_input('Hit ENTER to move') pr2_arm.go_cep_jtt(r_arm, p1) raw_input('Hit ENTER to go home') home = np.matrix([0.23, -0.6, -0.05]).T pr2_arm.go_cep_jtt(r_arm, home)
[ [ 1, 0, 0.0636, 0.0042, 0, 0.66, 0, 954, 0, 2, 0, 0, 954, 0, 0 ], [ 1, 0, 0.0678, 0.0042, 0, 0.66, 0.0476, 83, 0, 2, 0, 0, 83, 0, 0 ], [ 1, 0, 0.072, 0.0042, 0, 0.6...
[ "import numpy as np, math", "from threading import RLock, Timer", "import sys", "import roslib; roslib.load_manifest('hrl_pr2_lib')", "import roslib; roslib.load_manifest('hrl_pr2_lib')", "import tf", "import rospy", "import actionlib", "from actionlib_msgs.msg import GoalStatus", "from pr2_contro...
#!/usr/bin/env python # # Copyright (c) 2009, Georgia Tech Research Corporation # All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are met: # * Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # * Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # * Neither the name of the Georgia Tech Research Corporation nor the # names of its contributors may be used to endorse or promote products # derived from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. # #Author: Marc Killpack import roslib roslib.load_manifest('pr2_playpen') from UI_segment_object.srv import GetObject import rospy import sys import optparse p = optparse.OptionParser() p.add_option('--node', action='store', type='string', dest='node') p.add_option('--serv', action='store', type='string', dest='service') opt, args = p.parse_args() rospy.init_node(opt.node) rospy.wait_for_service(opt.service) pub_filtered_cloud = rospy.ServiceProxy(opt.service, GetObject) r = rospy.Rate(30) while not rospy.is_shutdown(): pub_filtered_cloud() r.sleep()
[ [ 1, 0, 0.5882, 0.0196, 0, 0.66, 0, 796, 0, 1, 0, 0, 796, 0, 0 ], [ 8, 0, 0.6078, 0.0196, 0, 0.66, 0.0714, 630, 3, 1, 0, 0, 0, 0, 1 ], [ 1, 0, 0.6275, 0.0196, 0, 0....
[ "import roslib", "roslib.load_manifest('pr2_playpen')", "from UI_segment_object.srv import GetObject", "import rospy", "import sys", "import optparse", "p = optparse.OptionParser()", "p.add_option('--node', action='store', type='string', dest='node')", "p.add_option('--serv', action='store', type='s...