rlightning.env.isaac_env

IsaacLab environment wrappers.

class rlightning.env.isaac_env.IsaacManagerBasedRLEnv(config: EnvConfig, worker_index: int | None = 0, preprocess_fn: Callable | None = <function default_env_preprocess_fn>, **kwargs: Any)

Bases: BaseEnv

IsaacLab manager-based RL environment wrapper.

close() → None

Close the environment and its simulation app.

get_action_space()

Retrive vectorized action space

Returns:

Action space

Return type:

Dict[str, gym.Space]

get_observation_space()

Retrieve vectorized observation space

Returns:

Observation space

Return type:

Dict[str, gym.Space]

init() → RSLEnvMeta

init the environment and return the environment meta information

reset(*args, **kwargs) → EnvRet

Reset the environment and return the initial EnvRet.

step(policy_resp: PolicyResponse) → EnvRet

Step the environment with the given policy response.

The returned EnvRet has items as a series of dict (mapping from agent ids to entries)

Parameters:

policy_resp (PolicyResponse) – The response from the policy, containing the action to take

Returns:

The return from the environment after taking the action

Return type:

EnvRet

property unwrapped: MockModule('gymnasium.Env')

Returns the base environment of the wrapper.

This will be the bare gymnasium.Env environment, underneath all layers of wrappers.

class rlightning.env.isaac_env.RSLEnvMeta(env_id, action_space, observation_space, num_envs, get_observations, num_actions)

Bases: tuple

action_space

Alias for field number 1

env_id

Alias for field number 0

get_observations

Alias for field number 4

num_actions

Alias for field number 5

num_envs

Alias for field number 3

observation_space

Alias for field number 2