jellyhoapache-2.0
BigYM TakeCups 20hz
A bimanual humanoid mobile manipulation dataset created with LeRobot containing 36 episodes of cup-taking tasks at 20 fps. The dataset includes proprioceptive state, environment state, actions, and multi-view RGB video observations.
Downloads30
Episodes36
Why This Matters for Physical AI
This dataset demonstrates bimanual humanoid mobile manipulation capabilities, providing multi-view observations and proprioceptive feedback essential for training embodied AI models on complex coordinated manipulation tasks.
Technical Profile
- Modalities
- rgbproprioception
- Robot Embodiments
- h1humanoidmobile_manipulator
- Action Space
- joint_positions
- Environment
- lab
- Task Types
- manipulationpick_and_place
- Episodes
- 36
- Data Format
- LeRobot
- Annotation Types
- reward_labelssuccess_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