XvKuoMingapache-2.0
so101_chess_wide
Chess moves and captures by an SO-101 arm with domain randomization across board sizes, materials, piece sets, camera placements, and lighting conditions. Scripted-expert demonstrations in MuJoCo with 7,521 episodes covering pick-and-place and capture tasks.
Downloads50
Episodes7,521
Why This Matters for Physical AI
Demonstrates how domain randomization across visual, environmental, and task parameters enables robust learning for manipulation tasks, with systematic evaluation on held-out distributions to measure generalization.
Technical Profile
- Modalities
- rgbproprioception
- Robot Embodiments
- SO-101
- Action Space
- joint_positions
- Environment
- simulation
- Task Types
- manipulationpick_and_place
- Episodes
- 7,521
- Data Format
- LeRobot
- Annotation Types
- language_instructionsaction_labels
- License
- apache-2.0
Access
Need custom rgb data?
Claru builds purpose-built datasets for simulation applications with dense human annotations and quality assurance.
Request a Sample Pack