Merge branch 'master' into deprecate_detpmp

# Conflicts:
#	README.md
#	alr_envs/alr/__init__.py

alr_envs/alr/classic_control/hole_reacher/hole_reacher.py

@@ -106,7 +106,7 @@ class HoleReacherEnv(BaseReacherDirectEnv):
             # self._tmp_hole_depth,
             self.end_effector - self._goal,
             self._steps
-        ])
+        ]).astype(np.float32)
 
     def _get_line_points(self, num_points_per_link=1):
         theta = self._joint_angles[:, None]

alr_envs/alr/classic_control/simple_reacher/simple_reacher.py

@@ -73,7 +73,7 @@ class SimpleReacherEnv(BaseReacherTorqueEnv):
             self._angle_velocity,
             self.end_effector - self._goal,
             self._steps
-        ])
+        ]).astype(np.float32)
 
     def _generate_goal(self):
 

alr_envs/alr/classic_control/viapoint_reacher/viapoint_reacher.py

@@ -110,7 +110,7 @@ class ViaPointReacherEnv(BaseReacherDirectEnv):
             self.end_effector - self._via_point,
             self.end_effector - self._goal,
             self._steps
-        ])
+        ]).astype(np.float32)
 
     def _check_collisions(self) -> bool:
         return self._check_self_collision()