metasim.cfg.robots

Base Class

BaseRobotCfg

Base configuration class for robots.

Classes

AntCfg	Base configuration class for robots.
FetchCfg	Base configuration class for robots.
FrankaCfg	Cfg for the Franka Emika Panda robot.
FrankaSlideUrdfCfg	Base configuration class for robots.
FrankaSlideUrdfSapienCfg	Base configuration class for robots.
FrankaUrdfCfg	Base configuration class for robots.
FrankaUrdfSapienCfg	Base configuration class for robots.
FrankaWithGripperExtensionCfg	Base configuration class for robots.
GoogleRobotStaticCfg	Base configuration class for robots.
H12WithoutHandCfg	Base configuration class for robots.
H1Cfg	Base configuration class for robots.
H1SimpleHandCfg	Base configuration class for robots.
IiwaCfg	Base configuration class for robots.
SawyerCfg	Base configuration class for robots.
SawyerMujocoCfg	Base configuration class for robots.
Ur5E2F85Cfg	Base configuration class for robots.
WalkerCfg	Base configuration class for robots.

BaseRobotCfg

class metasim.cfg.robots.BaseRobotCfg

Bases: ArticulationObjCfg

Base configuration class for robots.

num_joints: int

Number of joints in the robots.

actuators: dict[str, BaseActuatorCfg]

Actuators in the robots. The keys are the names of the actuators, and the values are the configurations of the actuators. The names should be consistent with the names in the asset file.

ee_body_name: str | None

Name of the end effector body, which should be consistent with the name in the asset file. This is used for end-effector control.

fix_base_link: bool

Whether to fix the base link. Default to True.

joint_limits: dict[str, tuple[float, float]]

Joint limits of the robots. The keys are the names of the joints, and the values are the limits of the joints. The names should be consistent with the names in the asset file.

name: str

Object name

default_position: tuple[float, float, float]

Default position of the object, default is (0.0, 0.0, 0.0)

default_orientation: tuple[float, float, float, float]

Default orientation of the object, default is (1.0, 0.0, 0.0, 0.0)

mesh_path: str | None

Path to the mesh file.

usd_path: str | None

Path to the USD file.

urdf_path: str | None

Path to the URDF file.

mjcf_path: str | None

Path to the MJCF file.

mjx_mjcf_path: str | None

Path to the MJCF file only used for MJX. If not specified, it will be the same as mjcf_path.

scale: float | tuple[float, float, float]

Object scaling (in scalar) for the object, default is 1.0

default_joint_positions: dict[str, float]

Default joint positions of the robots. The keys are the names of the joints, and the values are the default positions of the joints. The names should be consistent with the names in the asset file.

enabled_gravity: bool

Whether to enable gravity. Default to True. If False, the robot will not be affected by gravity.

enabled_self_collisions: bool

Whether to enable self collisions. Default to True. If False, the robot will not collide with itself.

isaacgym_flip_visual_attachments: bool

Whether to flip visual attachments when loading the URDF in IsaacGym. Default to True. For more details, see

• IsaacGym doc: https://docs.robotsfan.com/isaacgym/api/python/struct_py.html#isaacgym.gymapi.AssetOptions.flip_visual_attachments

collapse_fixed_joints: bool

Whether to collapse fixed joints when loading the URDF in IsaacGym or Genesis. Default to False. For more details, see

• IsaacGym doc: https://docs.robotsfan.com/isaacgym/api/python/struct_py.html#isaacgym.gymapi.AssetOptions.collapse_fixed_joints

• Genesis doc: https://genesis-world.readthedocs.io/en/latest/api_reference/options/morph/file_morph/urdf.html

gripper_open_q: list[float]

Joint positions of the gripper when the gripper is open. This is used for motion planning and retargetting using cuRobo.

gripper_close_q: list[float]

Joint positions of the gripper when the gripper is closed. This is used for motion planning and retargetting using cuRobo.

curobo_ref_cfg_name: str

Name of the configuration file for cuRobo. This is used for motion planning and retargetting using cuRobo.

curobo_tcp_rel_pos: tuple[float, float, float]

Relative position of the TCP to the end effector body link. This is used for motion planning and retargetting using cuRobo.

curobo_tcp_rel_rot: tuple[float, float, float]

Relative rotation of the TCP to the end effector body link. This is used for motion planning and retargetting using cuRobo.

control_type: dict[str, Literal['position', 'effort']]

Control type for each joint.

Warning

This is experimental and subject to change.