irlxrd2026apache-2.0
Objects to Bowl
A multi-task robotics dataset with 122 episodes of pick-and-place manipulation tasks collected using the SO-Follower robot, containing wrist and head camera views along with proprioceptive joint states.
Downloads208
Episodes122
Hours0.521
Why This Matters for Physical AI
This dataset provides multi-view demonstration data for learning pick-and-place manipulation skills with proprioceptive feedback, enabling training of visuomotor policies for robotic grasping and object manipulation tasks.
Technical Profile
- Modalities
- rgbproprioception
- Robot Embodiments
- SO-Follower
- 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
- pick_and_placemanipulation
- Episodes
- 122
- Total Hours
- 0.521
- Data Format
- LeRobot
- Annotation Types
- action_labels
- 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