zys10302026cc-by-4.0
Shape Pick-and-Place Shared-Autonomy Demonstrations
70 shared-autonomy-assisted demonstrations for pick-and-place tasks using human Cartesian translation and gripper control with assisted J6 yaw alignment. The dataset contains 14,212 frames from external RGB and wrist RGB cameras collected at 10 FPS.
Downloads8
Episodes70
Why This Matters for Physical AI
This dataset provides real-world demonstration data for training vision-language-action policies with shared autonomy, advancing research in imitation learning and robot learning from human guidance.
Technical Profile
- Modalities
- rgbproprioception
- Robot Embodiments
- RM-65
- Action Space
- end_effector_delta
- Environment
- lab
- Task Types
- pick_and_placemanipulation
- Episodes
- 70
- Data Format
- LeRobot
- Annotation Types
- language_instructions
- License
- cc-by-4.0
Access
Need custom rgb data?
Claru builds purpose-built datasets for lab applications with dense human annotations and quality assurance.
Request a Sample Pack