Is this page helpful?

Public API for module isaacsim.robot.policy.examples:#

Classes#

  • class IsaacLabPolicyController(mg.BaseController)

    • def init(self, model: PolicyModel, interface: BoundPolicy)

    • def reset(self, estimated_state: mg.RobotState, setpoint_state: mg.RobotState | None, t: float, **kwargs: object) -> bool

    • def forward(self, estimated_state: mg.RobotState, setpoint_state: mg.RobotState | None, t: float, **kwargs: object) -> mg.RobotState

    • def close(self)

  • class PolicyArtifact

    • model_path: str

    • env_config_path: str

    • backend: str

    • descriptor_path: str | None

    • model_sha256: str | None

    • def load_descriptor(self) -> Mapping

    • class def from_files(cls, model_path: str, env_config_path: str, descriptor_path: str | None = None) -> PolicyArtifact

    • class def from_bundle(cls, bundle_dir: str) -> PolicyArtifact

    • class def from_training_run(cls, run_dir: str) -> PolicyArtifact

  • class PolicyEnvConfig

    • def init(self, data: Mapping[str, Any], source_path: str | None = None)

    • class def from_file(cls, path: str) -> PolicyEnvConfig

    • [property] def timing(self) -> PhysicsTiming

    • [property] def spawn(self) -> SpawnProps

    • [property] def initial_root_pose(self) -> tuple[list[float] | None, list[float] | None]

    • def joint_properties(self, joint_names: Sequence[str]) -> JointProperties

    • def actuator_model_specs(self, joint_names: Sequence[str]) -> list[NewtonActuatorSpec]

  • class RobotPolicyRunner

    • def init(self, spec: PolicySpec)

    • [property] def articulation(self) -> Articulation | None

    • [property] def engine(self) -> str | None

    • [property] def physics_dt(self) -> float

    • [property] def decimation(self) -> int

    • def spawn(self) -> Articulation

    • def initialize(self)

    • def restart_from_default_state(self, command: Sequence[float] | None = None)

    • def step(self, dt: float, command: Sequence[float] | None = None)

    • def close(self)

  • class PolicySpec

    • engines: Mapping[str, PolicyArtifact]

    • usd_path: str | Mapping[str, str] | None

    • name: str | None

    • default_spawn_position: tuple[float, float, float] | None

    • default_spawn_orientation: tuple[float, float, float, float] | None

    • zero_targets_on_initialize: bool

    • binding: Callable[[PolicyEnvConfig, tuple[str, …]], PolicyBinding] | None

Functions#

  • def get_anymal_spec() -> PolicySpec

  • def get_cartpole_spec() -> PolicySpec

  • def get_franka_spec() -> PolicySpec

  • def get_go2_spec() -> PolicySpec

  • def get_h1_spec() -> PolicySpec

  • def get_spot_spec() -> PolicySpec

  • def make_franka_task_state_provider(cabinet: Articulation, env_config: PolicyEnvConfig) -> Callable[[Articulation], FrankaTaskStateProvider]

Public API for module isaacsim.robot.policy.examples.interactive.franka:#

Classes#

  • class FrankaExample(PolicySampleBase)

    • def init(self)

    • def setup_scene(self)

    • async def setup_post_load(self)

    • async def setup_pre_reset(self)

    • async def setup_post_reset(self)

    • def on_physics_step(self, dt: float, context: object)

    • def physics_cleanup(self)

Public API for module isaacsim.robot.policy.examples.interactive.go2:#

Classes#

  • class Go2Example(LocomotionPolicySample)

Public API for module isaacsim.robot.policy.examples.interactive.humanoid:#

Classes#

  • class HumanoidExample(LocomotionPolicySample)

Public API for module isaacsim.robot.policy.examples.interactive.quadruped:#

Classes#

  • class QuadrupedExample(LocomotionPolicySample)