def __init__(self, robot, render=False):
# print("WalkerBase::__init__ start")
self.camera_x = 0
self.walk_target_x = 1e3 # kilometer away
self.walk_target_y = 0
self.stateId = -1
MJCFBaseBulletEnv.__init__(self, robot, render)
def __init__(self, render=False):
self.robot = Kuka()
MJCFBaseBulletEnv.__init__(self, self.robot, render)
self._cam_dist = 1.2
self._cam_yaw = 90 #30
self._cam_roll = 0
self._cam_pitch = -70 #-30
if config.wtcheat:
self._render_width = 1
self._render_height = 1
else:
self._render_width = 320
self._render_height = 240
self._cam_pos = [0, 0, 0.3] #[0, 0, 0.4]
self.setCamera()
self.eyes = {}
self.reward_func = DefaultRewardFunc
self.robot.used_objects = ["table", "tomato", "mustard", "orange"]
self.set_eye("eye")
self.goal = Goal(retina=self.observation_space.spaces[
self.robot.ObsSpaces.GOAL].sample() * 0)
self.goals = None
self.goal_idx = -1
def __init__(self, robot, render=False, max_num_steps=1000):
# print("WalkerBase::__init__ start")
self.max_num_steps = max_num_steps
self._step_counter = 0
self.camera_x = 0
self.walk_target_x = 1e3 # kilometer away
self.walk_target_y = 0
self.stateId = -1
MJCFBaseBulletEnv.__init__(self, robot, render)
def __init__(self, robot, render=False):
# print("WalkerBase::__init__ start")
MJCFBaseBulletEnv.__init__(self, robot, render)
self.camera_x = 0
self.walk_target_x = 1e3 # kilometer away
self.walk_target_y = 0
self.stateId = -1
self._projectM = None
self._param_init_camera_width = 320
self._param_init_camera_height = 200
self._param_camera_distance = 2.0
def reset(self):
if (self.stateId >= 0):
# print("restoreState self.stateId:",self.stateId)
self._p.restoreState(self.stateId)
r = MJCFBaseBulletEnv.reset(self)
self._p.configureDebugVisualizer(pybullet.COV_ENABLE_RENDERING, 0)
self.parts, self.jdict, self.ordered_joints, self.robot_body = self.robot.addToScene(
self._p, self.stadium_scene.ground_plane_mjcf)
# self.ground_ids = set([(self.parts[f].bodies[self.parts[f].bodyIndex], self.parts[f].bodyPartIndex) for f in
# self.foot_ground_object_names])
self._p.configureDebugVisualizer(pybullet.COV_ENABLE_RENDERING, 1)
if (self.stateId < 0):
self.stateId = self._p.saveState()
#print("saving state self.stateId:",self.stateId)
self._p.setGravity(0, 0, -.5)
for _ in range(200):
self.robot.reset_position()
self.scene.global_step()
self.robot.reset_position_final()
self._p.setGravity(0, 0, -9.81)
r = self.robot.calc_state()
self.robot._initial_z = r[-1]
# for _ in range(20):
# self.scene.global_step()
# self.robot.reset_position_final()
# time.sleep(0.1)
# self.robot.reset_position()
# self.scene.global_step()
self.robot.initial_z = None
r = self.robot.calc_state()
return r
def reset(self):
if (self.stateId >= 0):
self._p.restoreState(self.stateId)
r = MJCFBaseBulletEnv.reset(self)
if (self.stateId < 0):
self.stateId = self._p.saveState()
return r
def __init__(self, render=False):
self.robot = Hand()
MJCFBaseBulletEnv.__init__(self, self.robot, render)
self._cam_dist = 1.2
self._cam_yaw = 30
self._cam_roll = 0
self._cam_pitch = -30
self._render_width = 320
self._render_height = 240
self._cam_pos = [0, 0, .4]
self.setCamera()
self.eyes = {}
self.reward_func = DefaultRewardFunc
self.robot.used_objects = []
def reset(self):
if (self.stateId >= 0):
#print("InvertedPendulumBulletEnv reset p.restoreState(",self.stateId,")")
self._p.restoreState(self.stateId)
r = MJCFBaseBulletEnv.reset(self)
if (self.stateId < 0):
self.stateId = self._p.saveState()
#print("InvertedPendulumBulletEnv reset self.stateId=",self.stateId)
return r
def __init__(self, render=False):
self.robot = Kuka()
MJCFBaseBulletEnv.__init__(self, self.robot, render)
self._cam_dist = 1.2
self._cam_yaw = 30
self._cam_roll = 0
self._cam_pitch = -30
self._render_width = 320
self._render_height = 240
self._cam_pos = [0,0,.4]
self.setCamera()
self.eyes = {}
self.reward_func = DefaultRewardFunc
self.robot.used_objects = ["table", "tomato", "mustard", "cube"]
self.set_eye("eye")
self.goal = Goal(retina=self.observation_space.spaces[
self.robot.ObsSpaces.GOAL].sample()*0)
# Set default goals dataset path
#
# The goals dataset is basically a list of real_robots.envs.env.Goal
# objects which are stored using :
#
# np.savez_compressed(
# "path.npy.npz",
# list_of_goals)
#
self.goals_dataset_path = os.path.join(
real_robots.getPackageDataPath(),
"goals_dataset.npy.npz")
self.goals = None
self.goal_idx = -1
def _reset(self):
if self.stateId >= 0:
# print("restoreState self.stateId:",self.stateId)
self._p.restoreState(self.stateId)
r = MJCFBaseBulletEnv._reset(self)
self._p.configureDebugVisualizer(pybullet.COV_ENABLE_RENDERING, 0)
self.parts, self.jdict, self.ordered_joints, self.robot_body = self.robot.addToScene(
self._p, self.stadium_scene.ground_plane_mjcf)
self.ground_ids = set([(
self.parts[f].bodies[self.parts[f].bodyIndex],
self.parts[f].bodyPartIndex,
) for f in self.foot_ground_object_names])
self._p.configureDebugVisualizer(pybullet.COV_ENABLE_RENDERING, 1)
return r
def reset(self):
if (self.stateId >= 0):
# print("restoreState self.stateId:",self.stateId)
self._p.restoreState(self.stateId)
r = MJCFBaseBulletEnv.reset(self)
self._p.configureDebugVisualizer(pybullet.COV_ENABLE_RENDERING, 0)
self.parts, self.jdict, self.ordered_joints, self.robot_body = self.robot.addToScene(
self._p, self.stadium_scene.ground_plane_mjcf)
self.ground_ids = set([(self.parts[f].bodies[self.parts[f].bodyIndex],
self.parts[f].bodyPartIndex) for f in self.foot_ground_object_names])
self._p.configureDebugVisualizer(pybullet.COV_ENABLE_RENDERING, 1)
if (self.stateId < 0):
self.stateId = self._p.saveState()
# print("saving state self.stateId:",self.stateId)
return r, {'v(m/s)':self.robot_body.speed(), 'pose(m)': self.robot.robot_body.pose().xyz(), 'rpy': self.robot.body_rpy, \
'ori': self.robot.robot_body.pose().orientation(), 'feet_contact': self.robot.feet_contact}
def reset(self):
if self.stateId >= 0:
# print("restoreState self.stateId:",self.stateId)
self._p.restoreState(self.stateId)
r = MJCFBaseBulletEnv.reset(self)
self._p.configureDebugVisualizer(pybullet.COV_ENABLE_RENDERING, 0)
(
self.parts,
self.jdict,
self.ordered_joints,
self.robot_body,
) = self.robot.addToScene(self._p, self.stadium_scene.terrain)
self._p.configureDebugVisualizer(pybullet.COV_ENABLE_RENDERING, 1)
if self.stateId < 0:
self.stateId = self._p.saveState()
# print("saving state self.stateId:",self.stateId)
return r
def _reset(self): if (self.stateId>=0): self._p.restoreState(self.stateId) rr = MJCFBaseBulletEnv._reset(self) self.beakerCam = BeakerCam(self._p) if(THROWBALLS and self.stateId < 0): self.frames_since_ball_thrown = 0 mass = .04 visualShapeId = -1 sphereRadius = 0.05 useMaximalCoordinates = 0 position = [0.3,0.3,0.2] colSphereId = self._p.createCollisionShape(p.GEOM_SPHERE,radius=sphereRadius) self.sphereUid = p.createMultiBody(mass,colSphereId,visualShapeId,position,useMaximalCoordinates=useMaximalCoordinates) self._throw_ball() if (self.stateId<0): self.stateId = self._p.saveState() return rr
def __init__(self):
self.robot = PendulumSwingup()
MJCFBaseBulletEnv.__init__(self, self.robot)
self.stateId = -1
def __init__(self, render=False, objects=3, action_type='joints',
additional_obs=True):
self.robot = Kuka(additional_obs, objects)
MJCFBaseBulletEnv.__init__(self, self.robot, render)
self.joints_space = self.robot.action_space
self.cartesian_space = spaces.Box(
low=np.array([-0.25, -0.5, 0.40, -1, -1, -1, -1]),
high=np.array([0.25, 0.5, 0.60, 1, 1, 1, 1]),
dtype=float)
self.macro_space = spaces.Box(
low=np.array([[-0.25, -0.5], [-0.25, -0.5]]),
high=np.array([[0.05, 0.5], [0.05, 0.5]]),
dtype=float)
self.gripper_space = spaces.Box(low=0,
high=np.pi/2, shape=(2,), dtype=float)
if action_type == 'joints':
self.action_space = spaces.Dict({
"joint_command": self.joints_space,
"render": spaces.MultiBinary(1)})
self.step = self.step_joints
elif action_type == 'cartesian':
self.action_space = spaces.Dict({
"cartesian_command": self.cartesian_space,
"gripper_command": self.gripper_space,
"render": spaces.MultiBinary(1)})
self.step = self.step_cartesian
self.requested_coords = None
self.requested_orient = None
self.last_ik = None
elif action_type == 'macro_action':
self.action_space = spaces.Dict({
"macro_action": self.macro_space,
"render": spaces.MultiBinary(1)})
self.step = self.step_macro
self.requested_action = None
else:
raise ValueError("action_type must be one 'joints', 'cartesian' "
"or 'macro_action'")
self._cam_dist = 1.2
self._cam_yaw = 30
self._cam_roll = 0
self._cam_pitch = -30
self._render_width = 320
self._render_height = 240
self._cam_pos = [0, 0, .4]
self.setCamera()
self.eyes = {}
self.reward_func = DefaultRewardFunc
self.set_eye("eye")
self.goal = Goal(retina=self.observation_space.spaces[
self.robot.ObsSpaces.GOAL].sample()*0)
# Set default goals dataset path
#
# The goals dataset is basically a list of real_robots.envs.env.Goal
# objects which are stored using :
#
# np.savez_compressed(
# "path.npy.npz",
# list_of_goals)
#
self.goals_dataset_path = os.path.join(
real_robots.getPackageDataPath(),
"goals_dataset.npy.npz")
self.goals = None
self.goal_idx = -1
self.no_retina = self.observation_space.spaces[
self.robot.ObsSpaces.RETINA].sample()*0
self.no_depth = self.observation_space.spaces[
self.robot.ObsSpaces.DEPTH].sample()*0
if additional_obs:
self.get_observation = self.get_observation_extended
self.no_mask = self.observation_space.spaces[
self.robot.ObsSpaces.MASK].sample()*0
def __init__(self):
self.robot = InvertedPendulumModified()
MJCFBaseBulletEnv.__init__(self, self.robot)
self.stateId = -1
def __init__(self, render=False):
self.robot = FixedReacher()
MJCFBaseBulletEnv.__init__(self, self.robot, render)
def __init__(self): self.robot = BeakerBot() MJCFBaseBulletEnv.__init__(self, self.robot) self.stateId=-1