ghkim-rlwrldapache-2.0
shell_game — delta EEF, global instruction
A robotics manipulation dataset where a robotic arm lifts cups to find a hidden red block after manual shuffling. Contains 45 episodes with delta end-effector actions and global language instructions.
Downloads23
Episodes45
Why This Matters for Physical AI
This dataset demonstrates real robot manipulation with natural language instructions and delta end-effector control, providing a benchmark for training policies that combine visual perception with linguistic task specification.
Technical Profile
- Modalities
- rgblanguage
- Robot Embodiments
- robotic arm
- Action Space
- end_effector_delta
- Environment
- lab
- Task Types
- manipulationobject_search
- Episodes
- 45
- Data Format
- mp4
- Annotation Types
- language_instructions
- 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