oxe-auge2025cc-by-4.0
language_table_train_220000_225000_augmented
An augmented subset of the Language Table dataset containing 5,000 episodes with robot visual and proprioceptive observations augmented across 8 different robot embodiments (Google Robot, Jaco, Kinova3, KUKA IIWA, Panda, Sawyer, UR5e) with natural language task instructions.
Downloads67
Episodes5000
Why This Matters for Physical AI
This dataset enables cross-embodiment policy learning by providing augmented visual and proprioceptive data across multiple robot platforms with consistent natural language instructions, addressing the challenge of generalizing manipulation skills across different robot morphologies.
Technical Profile
- Modalities
- rgbproprioceptionlanguage
- Robot Embodiments
- Google RobotJacoKinova3KUKA IIWAPandaSawyerUR5e
- Environment
- lab
- Task Types
- manipulation
- Episodes
- 5000
- Data Format
- parquet
- 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