def __init__(self):
super(RLreconSimpleV2Env, self).__init__()
rospy.init_node('gym_RLrecon_simple_node', anonymous=False)
world_bounding_box = math_utils.BoundingBox(
[-20, -20, 0],
[+20, +20, +20],
)
score_bounding_box = math_utils.BoundingBox([-3, -3, -0.5],
[+3, +3, +5])
engine = DummyEngine()
self._environment = SimpleV2Environment(
world_bounding_box,
engine=engine,
random_reset=True,
clear_size=6.0,
filter_depth_map=False,
score_bounding_box=score_bounding_box)
self.action_space = spaces.Discrete(self._environment.num_actions())
# self.observation_space = spaces.Box(
# 0.0,
# 1.0,
# shape=self._environment.observation_shape()
# )
self._obs_level = 3
self._obs_size_x = 16
self._obs_size_y = self._obs_size_x
self._obs_size_z = self._obs_size_x
self.observation_space = spaces.Box(0.0,
1.0,
shape=(self._obs_size_x,
self._obs_size_y,
self._obs_size_z, 1))
self.observation_space = spaces.Box(-1000, +1000, shape=(14, ))
def __init__(self):
super(RLreconDummyEnv, self).__init__()
rospy.init_node('gym_RLrecon_dummy_node', anonymous=False)
bounding_box = math_utils.BoundingBox([-10, -10, 0], [+10, +10, +10])
engine = DummyEngine()
self._environment = SimpleEnvironment(bounding_box,
engine=engine,
random_reset=True,
clear_size=6.0)
self.action_space = spaces.Discrete(self._environment.num_actions())
self.observation_space = spaces.Box(-1000, +1000, shape=(2, ))
def __init__(self, environment_class):
super(RLreconEnvWrapper, self).__init__()
rospy.init_node('gym_RLrecon_env_wrapper', anonymous=False)
world_bounding_box = math_utils.BoundingBox(
[-20, -20, 0],
[+20, +20, +20],
)
score_bounding_box = math_utils.BoundingBox([-3, -3, -0.5],
[+3, +3, +5])
engine = DummyEngine()
self._environment = environment_class(
world_bounding_box,
engine=engine,
random_reset=True,
clear_size=6.0,
filter_depth_map=False,
score_bounding_box=score_bounding_box)
self.action_space = spaces.Discrete(
self._environment.get_num_of_actions())
# self.observation_space = spaces.Box(
# 0.0,
# 1.0,
# shape=self._environment.observation_shape()
# )
# self._obs_level = 3
# self._obs_size_x = 16
# self._obs_level = 4
# self._obs_size_x = 8
# self._obs_size_y = self._obs_size_x
# self._obs_size_z = self._obs_size_x
# self.observation_space = spaces.Box(
# 0.0,
# 1.0,
# shape=(self._obs_size_x, self._obs_size_y, self._obs_size_z, 2)
# )
# self.observation_space = spaces.Box(
# -1000,
# +1000,
# shape=(14,)
# )
spaces_tuple = []
for obs_shape in self._environment.get_observation_shapes():
spaces_tuple.append(
spaces.Box(-np.finfo(np.float32).max,
+np.finfo(np.float32).max,
shape=obs_shape))
self.observation_space = spaces.Tuple(spaces_tuple)