salehin212026cc-by-nc-4.0
DecisionFacts Physical AI Dataset — SO-101 Robotic Arm Teleoperation
A curated collection of real-world teleoperation data captured on the SO-101 robotic arm, with synchronized multi-modal capture across dual cameras, proprioceptive state, and control signals at 30 fps, designed to support training of imitation-learning policies, Vision-Language-Action models, and world models.
Downloads85
Episodes43
Hours0.4426
Why This Matters for Physical AI
This dataset provides high-quality, standardized teleoperated demonstrations on a low-cost, reproducible robotic platform to support training of vision-language-action models, world models, and imitation-learning policies for embodied AI research.
Technical Profile
- Modalities
- rgbproprioception
- Robot Embodiments
- SO-101
- Action Space
- 6-DoF: shoulder_pan.pos, shoulder_lift.pos, elbow_flex.pos, wrist_flex.pos, wrist_roll.pos, gripper.pos
- Environment
- lab
- Task Types
- manipulationgrasping
- Episodes
- 43
- Total Hours
- 0.4426
- Data Format
- LeRobot
- Annotation Types
- language_instructionsaction_labelstask_outcome_labels
- License
- cc-by-nc-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