Skip to main content
Embodiment tags are used to identify the robot embodiment in the data. They enable cross-embodiment training by providing consistent identifiers across datasets.

Naming convention

Embodiment tags follow the pattern <dataset>_<robot_name>. If using multiple datasets with the same robot (e.g., sim GR1 and real GR1), you can drop the dataset name and use only the robot name.

EmbodimentTag

Enum class for robot embodiment identifiers.

Pretrain embodiment tags

These embodiments are used during the pretraining phase:

ROBOCASA_PANDA_OMRON

The RoboCasa Panda robot with Omron mobile base.

GR1

The Fourier GR1 robot.

Posttrain embodiment tags

These embodiments are pre-registered for post-training:

UNITREE_G1

The Unitree G1 robot.

LIBERO_PANDA

The Libero Panda robot.

OXE_GOOGLE

The Open-X-Embodiment Google robot.

OXE_WIDOWX

The Open-X-Embodiment WidowX robot.

OXE_DROID

The Open-X-Embodiment DROID robot with relative joint position actions.

BEHAVIOR_R1_PRO

The Behavior R1 Pro robot.

Custom embodiments

NEW_EMBODIMENT

Placeholder tag for any new embodiment during post-training.

Usage