sreetz-nvapache-2.0
LeRobot SO-100 Pick-and-Place Dataset
A robotics dataset created with LeRobot containing 71 episodes of pick-and-place tasks performed by an SO-100 follower robot, with RGB video from ego and external cameras plus proprioceptive state information.
Downloads95
Episodes71
Why This Matters for Physical AI
This dataset provides multi-view RGB video and joint proprioceptive data for training visuomotor policies on robotic manipulation tasks using the LeRobot framework.
Technical Profile
- Modalities
- rgbproprioception
- Robot Embodiments
- so101_follower
- Action Space
- joint_positions
- Environment
- lab
- Task Types
- pick_and_place
- Episodes
- 71
- Data Format
- parquet
- License
- apache-2.0
Access
Need custom rgb data?
Claru builds purpose-built datasets for lab applications with dense human annotations and quality assurance.
Request a Sample Pack