mulliganMIT
real-marker-d2-c02-dagger-mulligan
A robotics dataset from the Mulligan project containing 50 episodes of marker-based manipulation tasks with multi-camera observations. Part of the Mulligan policy arena release for training and evaluation of robotic policies.
Downloads201
Episodes50
Why This Matters for Physical AI
This dataset contributes to physical AI research by providing multi-view RGB observations and trajectory data from real robotic manipulation tasks, enabling training and evaluation of visuomotor policies through the Mulligan framework.
Technical Profile
- Modalities
- rgb
- Environment
- lab
- Task Types
- manipulation
- Episodes
- 50
- Data Format
- parquet
- Annotation Types
- language_instructions
- License
- MIT
Access
Need custom rgb data?
Claru builds purpose-built datasets for lab applications with dense human annotations and quality assurance.
Request a Sample Pack