visual_robust_libero — Experiment 08: static embodiment supervision
A cross-embodiment robotic manipulation dataset built on LIBERO with 24 embodiments (6 robots × 4 grippers) designed to test whether supervising policies on static embodiment information improves task transfer. Includes action demonstrations, synthetic replay data, and visual question-answering annotations.
Why This Matters for Physical AI
This dataset directly addresses cross-embodiment transfer learning by providing controlled comparisons of identical end-effector trajectories executed across 24 different robot-gripper combinations, enabling research on how static embodiment supervision and coordinate frame choices affect generalization across robot morphologies.
Technical Profile
- Modalities
- rgbdepthproprioceptionlanguage
- Robot Embodiments
- PandaIIWAUR5eJacoKinova3Sawyer
- Action Space
- end_effector_delta
- Environment
- simulation
- Task Types
- manipulationreachingpushing
- Episodes
- 438
- Data Format
- LeRobot
- Annotation Types
- language_instructionsaction_labelssegmentationembodiment_metadata
- License
- mit
Access
Need custom rgb data?
Claru builds purpose-built datasets for simulation applications with dense human annotations and quality assurance.
Request a Sample Pack