Closed maxspahn closed 1 year ago
Observations and actions can already be flattened using a similar method as in https://github.com/maxspahn/gym_envs_urdf/blob/be7532ae35675c5a2fd8c0d1782e8dbfd684e446/urdfenvs/urdf_common/urdf_env.py#L122-L130
Only now, it must be done for gym dicts.
This is required for RL algorithms to work, see #170.
Observations and actions can already be flattened using a similar method as in https://github.com/maxspahn/gym_envs_urdf/blob/be7532ae35675c5a2fd8c0d1782e8dbfd684e446/urdfenvs/urdf_common/urdf_env.py#L122-L130
Only now, it must be done for gym dicts.
This is required for RL algorithms to work, see #170.