added DMC Reacher, cartpole, reach_size; removed BBO

alr_envs/dmc/manipulation/reach/reach_mp_wrapper.py

@@ -0,0 +1,38 @@
+from typing import Tuple, Union
+
+import numpy as np
+
+from mp_env_api.interface_wrappers.mp_env_wrapper import MPEnvWrapper
+
+
+class DMCReachSiteMPWrapper(MPEnvWrapper):
+
+    @property
+    def active_obs(self):
+        # Joint and target positions are randomized, velocities are always set to 0.
+        return np.hstack([
+            [True] * 3,  # target position
+            [True] * 12,  # sin/cos arm joint position
+            [True] * 6,  # arm joint torques
+            [False] * 6,  # arm joint velocities
+            [True] * 3,  # sin/cos hand joint position
+            [False] * 3,  # hand joint velocities
+            [True] * 3,  # hand pinch site position
+            [True] * 9,  # pinch site rmat
+        ])
+
+    @property
+    def current_pos(self) -> Union[float, int, np.ndarray]:
+        return self.env.physics.named.data.qpos[:]
+
+    @property
+    def current_vel(self) -> Union[float, int, np.ndarray, Tuple]:
+        return self.env.physics.named.data.qvel[:]
+
+    @property
+    def goal_pos(self) -> Union[float, int, np.ndarray, Tuple]:
+        raise ValueError("Goal position is not available and has to be learnt based on the environment.")
+
+    @property
+    def dt(self) -> Union[float, int]:
+        return self.env.dt

alr_envs/dmc/suite/ball_in_cup/ball_in_cup_mp_wrapper.py

@@ -19,11 +19,11 @@ class DMCBallInCupMPWrapper(MPEnvWrapper):
 
     @property
     def current_pos(self) -> Union[float, int, np.ndarray]:
-        return np.hstack([self.physics.named.data.qpos['cup_x'], self.physics.named.data.qpos['cup_z']])
+        return np.hstack([self.env.physics.named.data.qpos['cup_x'], self.env.physics.named.data.qpos['cup_z']])
 
     @property
     def current_vel(self) -> Union[float, int, np.ndarray, Tuple]:
-        return np.hstack([self.physics.named.data.qvel['cup_x'], self.physics.named.data.qvel['cup_z']])
+        return np.hstack([self.env.physics.named.data.qvel['cup_x'], self.env.physics.named.data.qvel['cup_z']])
 
     @property
     def goal_pos(self) -> Union[float, int, np.ndarray, Tuple]:

alr_envs/dmc/suite/reacher/reacher_mp_wrapper.py

@@ -0,0 +1,33 @@
+from typing import Tuple, Union
+
+import numpy as np
+
+from mp_env_api.interface_wrappers.mp_env_wrapper import MPEnvWrapper
+
+
+class DMCReacherMPWrapper(MPEnvWrapper):
+
+    @property
+    def active_obs(self):
+        # Joint and target positions are randomized, velocities are always set to 0.
+        return np.hstack([
+            [True] * 2,  # joint position
+            [True] * 2,  # target position
+            [False] * 2,  # joint velocity
+        ])
+
+    @property
+    def current_pos(self) -> Union[float, int, np.ndarray]:
+        return self.env.physics.named.data.qpos[:]
+
+    @property
+    def current_vel(self) -> Union[float, int, np.ndarray, Tuple]:
+        return self.env.physics.named.data.qvel[:]
+
+    @property
+    def goal_pos(self) -> Union[float, int, np.ndarray, Tuple]:
+        raise ValueError("Goal position is not available and has to be learnt based on the environment.")
+
+    @property
+    def dt(self) -> Union[float, int]:
+        return self.env.dt