def testInitWithCommunicator(self):
env = RTRLBaseEnv({'generic': {'Communicator': MockCommunicator, 'kwargs': {}}}, 2, 3)
self.assertFalse(env._running)
self.assertEqual(len(env._all_comms), 1)
self.assertEqual(env._action_buffer.array_len, 2)
self.assertEqual(env._sensation_buffer.array_len, 5)
def testStartWithCommunicator(self):
env = RTRLBaseEnv({'generic': {'Communicator': MockCommunicator, 'kwargs': {}}}, 2, 3, run_mode='singlethread')
env.start()
time.sleep(0.5)
self.assertTrue(env._running)
self.assertTrue(env._all_comms['generic'].is_alive())
env.close()
self.assertFalse(env._running)
self.assertFalse(env._all_comms['generic'].is_alive())
def testStartMultiprocess(self):
env = RTRLBaseEnv({}, 2, 3, run_mode='multiprocess')
env.start()
self.assertTrue(env._running)
time.sleep(0.5)
self.assertTrue(env._polling_loop.is_alive())
env.close()
self.assertFalse(env._running)
self.assertFalse(env._polling_loop.is_alive())
def testNotImplementedError(self):
env = RTRLBaseEnv({}, 2, 3, run_mode='singlethread')
env.start()
with self.assertRaises(NotImplementedError):
env.step(0)
env.close()
def testStartSingalthread(self):
env = RTRLBaseEnv({}, 2, 3, run_mode='singlethread')
env.start()
self.assertTrue(env._running)
env.close()
self.assertFalse(env._running)
def testInit(self):
env = RTRLBaseEnv({}, 2, 3)
self.assertFalse(env._running)
self.assertEqual(env._action_buffer.array_len, 2)
self.assertEqual(env._sensation_buffer.array_len, 5)
def __init__(
self,
setup,
camera_hosts=('localhost', ),
camera_ports=(5000, ),
camera_res=(3, 480, 640),
host=None,
q_start_queue=None,
q_target=None,
dof=6,
control_type='position',
derivative_type='none',
target_type='position',
reset_type='random',
reward_type='linear',
deriv_action_max=10,
first_deriv_max=10, # used only with second derivative control
vel_penalty=0,
obs_history=1,
actuation_sync_period=1,
episode_length_time=None,
episode_length_step=None,
rllab_box=False,
servoj_t=ur_utils.COMMANDS['SERVOJ']['default']['t'],
servoj_gain=ur_utils.COMMANDS['SERVOJ']['default']['gain'],
speedj_a=ur_utils.COMMANDS['SPEEDJ']['default']['a'],
speedj_t_min=ur_utils.COMMANDS['SPEEDJ']['default']['t_min'],
movej_t=2, # used for resetting
accel_max=None,
speed_max=None,
dt=0.008,
delay=0.0, # to simulate extra delay in the system
**kwargs):
assert len(camera_hosts) == len(camera_ports)
self.num_cameras = len(camera_hosts)
self.camera_res = camera_res
self.camera_dim = int(np.product(camera_res))
self.buffer_len = obs_history
self.q_start_queue = q_start_queue
self.q_target = q_target
# Setup camera communicators and buffer
communicator_setups = {}
self._camera_images_ = {}
for idx, (camera_host,
camera_port) in enumerate(zip(camera_hosts, camera_ports)):
communicator_setups['Camera_{}'.format(idx)] = {
'Communicator': RealSenseCommunicator,
# have to read in this number of packets everytime to support
# all operations
'num_sensor_packets': self.buffer_len,
'kwargs': {
'host': camera_host,
'port': camera_port,
'num_channels': camera_res[0],
'height': camera_res[1],
'width': camera_res[2]
}
}
self._camera_images_['Camera_{}'.format(idx)] = np.frombuffer(
Array('f', self.camera_dim).get_obj(), dtype='float32')
print("Setup Reacher Environment")
# Setup UR environment
super(ReacherEnvWithRealSense, self).__init__(
setup,
host=host,
dof=dof,
control_type=control_type,
derivative_type=derivative_type,
target_type=target_type,
reset_type=reset_type,
reward_type=reward_type,
deriv_action_max=deriv_action_max,
first_deriv_max=
first_deriv_max, # used only with second derivative control
vel_penalty=vel_penalty,
obs_history=obs_history,
actuation_sync_period=actuation_sync_period,
episode_length_time=episode_length_time,
episode_length_step=episode_length_step,
rllab_box=rllab_box,
servoj_t=servoj_t,
servoj_gain=servoj_gain,
speedj_a=speedj_a,
speedj_t_min=speedj_t_min,
movej_t=movej_t, # used for resetting
accel_max=accel_max,
speed_max=speed_max,
dt=dt,
delay=delay, # to simulate extra delay in the system
communicator_setups=communicator_setups,
**kwargs)
# Update the observation space from ReacherEnv to include camera
self.joint_dim = int(np.product(self._observation_space.shape))
self.input_dim = self.joint_dim + self.num_cameras * self.camera_dim
if rllab_box:
from rllab.spaces import Box as RlBox # use this for rllab TRPO
Box = RlBox
else:
from gym.spaces import Box as GymBox # use this for baselines algos
Box = GymBox
self._observation_space = Box(low=-np.concatenate((np.zeros(
self.num_cameras * self.camera_dim), self._observation_space.low)),
high=np.concatenate(
(np.ones(self.num_cameras *
self.camera_dim),
self._observation_space.high)))
print("Communicators Setup")
RTRLBaseEnv.__init__(self,
communicator_setups=communicator_setups,
action_dim=len(self.action_space.low),
observation_dim=len(self.observation_space.low),
dt=dt,
**kwargs)
self._obs_ = np.zeros(self.input_dim)
