mani_skill.utils.building.mjcf_loader

Classes

MJCFLoader	Wrapper for the SAPIEN MJCF Loader to support easy parallelization
ParsedMJCFData	dict() -> new empty dictionary

Module Contents

class mani_skill.utils.building.mjcf_loader.MJCFLoader(ignore_classes=['motor'], visual_groups=[0, 2])

Bases: mani_skill.utils.building._mjcf_loader.MJCFLoader

Wrapper for the SAPIEN MJCF Loader to support easy parallelization

load(mjcf_file, package_dir=None, name=None, scene_idxs=None)

Parameters:

• urdf_file – filename for URDL file

• srdf_file – SRDF for urdf_file. If srdf_file is None, it defaults to the “.srdf” file with the same as the urdf file

• package_dir – base directory used to resolve asset files in the URDF file. If an asset path starts with “package://”, “package://” is simply removed from the file name

• name (str) – name of the created articulation

• scene_idxs (list[int]) – the ids of the scenes to build the objects in

• mjcf_file (str) –

Returns:

returns a single Articulation loaded from the URDF file. It throws an error if multiple objects exists

Return type:

mani_skill.utils.structs.Articulation

parse(mjcf_file, package_dir=None)

Parses a given MJCF file into articulation builders and actor builders and sensor configs

Return type:

ParsedMJCFData

disable_self_collisions: bool = False

name: str = None

scene: mani_skill.envs.scene.ManiSkillScene

class mani_skill.utils.building.mjcf_loader.ParsedMJCFData

Bases: TypedDict

dict() -> new empty dictionary dict(mapping) -> new dictionary initialized from a mapping object’s

(key, value) pairs

dict(iterable) -> new dictionary initialized as if via:

d = {} for k, v in iterable:

d[k] = v

dict(**kwargs) -> new dictionary initialized with the name=value pairs

in the keyword argument list. For example: dict(one=1, two=2)

actor_builders: list[mani_skill.utils.building.actor_builder.ActorBuilder]

articulation_builders: list[mani_skill.utils.building.articulation_builder.ArticulationBuilder]

cameras: list[Any]