Merge pull request #64 from ALRhub/Add-ProDMP-envs

Add prodmp metaworld envs

fancy_gym/black_box/black_box_wrapper.py

@@ -62,7 +62,6 @@ class BlackBoxWrapper(gym.ObservationWrapper):
         self.return_context_observation = not (learn_sub_trajectories or self.do_replanning)
         self.traj_gen_action_space = self._get_traj_gen_action_space()
         self.action_space = self._get_action_space()
-
         self.observation_space = self._get_observation_space()
 
         # rendering
@@ -95,23 +94,16 @@ class BlackBoxWrapper(gym.ObservationWrapper):
         clipped_params = np.clip(action, self.traj_gen_action_space.low, self.traj_gen_action_space.high)
         self.traj_gen.set_params(clipped_params)
         init_time = np.array(0 if not self.do_replanning else self.current_traj_steps * self.dt)
-        # TODO we could think about initializing with the previous desired value in order to have a smooth transition
-        #  at least from the planning point of view.
 
         condition_pos = self.condition_pos if self.condition_pos is not None else self.current_pos
         condition_vel = self.condition_vel if self.condition_vel is not None else self.current_vel
 
         self.traj_gen.set_initial_conditions(init_time, condition_pos, condition_vel)
         self.traj_gen.set_duration(duration, self.dt)
-        # traj_dict = self.traj_gen.get_trajs(get_pos=True, get_vel=True)
+
         position = get_numpy(self.traj_gen.get_traj_pos())
         velocity = get_numpy(self.traj_gen.get_traj_vel())
 
-        # if self.do_replanning:
-        #     # Remove first part of trajectory as this is already over
-        #     position = position[self.current_traj_steps:]
-        #     velocity = velocity[self.current_traj_steps:]
-
         return position, velocity
 
     def _get_traj_gen_action_space(self):
@@ -182,12 +174,13 @@ class BlackBoxWrapper(gym.ObservationWrapper):
                                                  t + 1 + self.current_traj_steps)
                         and self.plan_steps < self.max_planning_times):
 
-                self.condition_pos = pos if self.condition_on_desired else None
+                if self.condition_on_desired:
-                self.condition_vel = vel if self.condition_on_desired else None
+                    self.condition_pos = pos
+                    self.condition_vel = vel
 
                 break
 
-        infos.update({k: v[:t+1] for k, v in infos.items()})
+        infos.update({k: v[:t + 1] for k, v in infos.items()})
         self.current_traj_steps += t + 1
 
         if self.verbose >= 2:
@@ -210,6 +203,6 @@ class BlackBoxWrapper(gym.ObservationWrapper):
         self.current_traj_steps = 0
         self.plan_steps = 0
         self.traj_gen.reset()
-        self.condition_vel = None
         self.condition_pos = None
+        self.condition_vel = None
         return super(BlackBoxWrapper, self).reset()

fancy_gym/meta/__init__.py

@@ -13,7 +13,8 @@ DEFAULT_BB_DICT_ProMP = {
     "name": 'EnvName',
     "wrappers": [],
     "trajectory_generator_kwargs": {
-        'trajectory_generator_type': 'promp'
+        'trajectory_generator_type': 'promp',
+        'weights_scale': 10,
     },
     "phase_generator_kwargs": {
         'phase_generator_type': 'linear'
@@ -25,6 +26,9 @@ DEFAULT_BB_DICT_ProMP = {
         'basis_generator_type': 'zero_rbf',
         'num_basis': 5,
         'num_basis_zero_start': 1
+    },
+    'black_box_kwargs': {
+        'condition_on_desired': False,
     }
 }
 
@@ -32,22 +36,28 @@ DEFAULT_BB_DICT_ProDMP = {
     "name": 'EnvName',
     "wrappers": [],
     "trajectory_generator_kwargs": {
-        'trajectory_generator_type': 'prodmp'
+        'trajectory_generator_type': 'prodmp',
+        'auto_scale_basis': True,
+        'weights_scale': 10,
+        # 'goal_scale': 0.,
+        'disable_goal': True,
     },
     "phase_generator_kwargs": {
-        'phase_generator_type': 'exp'
+        'phase_generator_type': 'exp',
+        # 'alpha_phase' : 3,
     },
     "controller_kwargs": {
         'controller_type': 'metaworld',
     },
     "basis_generator_kwargs": {
         'basis_generator_type': 'prodmp',
-        'num_basis': 5
+        'num_basis': 3,
+        'alpha': 10
     },
-    "black_box_kwargs": {
+    'black_box_kwargs': {
-        'replanning_schedule': None,
+        'condition_on_desired': False,
-        'max_planning_times': None,
     }
+
 }
 
 _goal_change_envs = ["assembly-v2", "pick-out-of-hole-v2", "plate-slide-v2", "plate-slide-back-v2",
@@ -152,7 +162,6 @@ for _task in _goal_and_object_change_envs:
     )
     ALL_METAWORLD_MOVEMENT_PRIMITIVE_ENVIRONMENTS["ProDMP"].append(_env_id)
 
-
 _goal_and_endeffector_change_envs = ["basketball-v2"]
 for _task in _goal_and_endeffector_change_envs:
     task_id_split = _task.split("-")

fancy_gym/meta/base_metaworld_mp_wrapper.py

@@ -9,12 +9,9 @@ class BaseMetaworldMPWrapper(RawInterfaceWrapper):
     @property
     def current_pos(self) -> Union[float, int, np.ndarray]:
         r_close = self.env.data.get_joint_qpos("r_close")
-        # TODO check if this is correct
-        # return np.hstack([self.env.data.get_body_xpos('hand').flatten() / self.env.action_scale, r_close])
         return np.hstack([self.env.data.mocap_pos.flatten() / self.env.action_scale, r_close])
 
     @property
     def current_vel(self) -> Union[float, int, np.ndarray, Tuple]:
-        # TODO check if this is correct
         return np.zeros(4, )
         # raise NotImplementedError("Velocity cannot be retrieved.")

test/test_metaworld_envs.py

@@ -9,7 +9,6 @@ from test.utils import run_env, run_env_determinism
 METAWORLD_IDS = [f'metaworld:{env.split("-goal-observable")[0]}' for env, _ in
                  ALL_V2_ENVIRONMENTS_GOAL_OBSERVABLE.items()]
 METAWORLD_MP_IDS = chain(*fancy_gym.ALL_METAWORLD_MOVEMENT_PRIMITIVE_ENVIRONMENTS.values())
-print(METAWORLD_MP_IDS)
 SEED = 1