30) > 15 or abs(robot.getAngle("r" + joint) - 30) > 15:
print "!!!!!!!!!! Joint " + joint + " in wrong position " + str(
robot.getAngle("r" + joint) -
30) + " . Please do not break my arm! !!!!!!!!!!!!"
quit()
else:
print ""
else:
print "!!!!!!!!!! Joint " + joint + " in wrong position . Please do not break my arm! !!!!!!!!!!!!"
quit()
# set the robot to be compliant
#robot.disableTorqueAll()
mover_path = "../../../../moves_and_positions/"
mov = Mover.Mover(robot, stiff_off=False)
fe = faceExpression()
fe.sendFaceExpression("happiness")
#time.sleep(1)
# build the neural network
print("Building network")
network = build_net()
print("Loading network from files")
# And load parameters from file
# used for evaqluation : model_selflearning-CORE1-0.0701329990989.npz
with np.load('model_selflearning-MERGED-0.0455165514722.npz') as f:
# Very simple example to use the mover class
# Toggle the right arm between two position and open and close the hands
# (Leads to a dice throw, if you put the dice in the robots hand)
import logging
import time
from nicomotion import Motion, Mover
logging.basicConfig(level=logging.WARNING)
mover_path = "../../../moves_and_positions/"
robot = Motion.Motion("../../../json/nico_humanoid_upper.json", vrep=False)
mov = Mover.Mover(robot, stiff_off=True)
key = "c"
while key != "q":
# mov.play_movement(mover_path+"mov_from_table_to_get.csv",
# subsetfname=mover_path+"subset_left_arm_and_head.csv",move_speed=0.01)
mov.move_file_position(mover_path + "pos_from_table_to_get.csv",
subsetfname=(
mover_path + "subset_left_arm_and_head.csv"),
move_speed=0.05)
robot.openHand("LHand")
raw_input()
robot.closeHand("LHand", 1.0, 0.5)
time.sleep(2)
# mov.play_movement(mover_path+"mov_from_get_throw.csv",subsetfname=mover_path+"subset_left_arm.csv",move_speed=0.03)
ttw = mov.move_file_position(mover_path + "pos_from_get_throw.csv",