mulligan2026mit
sim-square-broad-c01-dagger-sobol-free-cf
A simulated robotic manipulation dataset for the sim-square-broad task, part of the Mulligan project. Contains 100 episodes of training data collected with a sobol-free-cf variant using dagger methodology.
Downloads59
Episodes100
Why This Matters for Physical AI
This dataset contributes to physical AI research by providing simulated manipulation trajectories with dagger-based data collection, enabling the study of policy learning and imitation in controlled robotic environments.
Technical Profile
- Modalities
- proprioception
- Environment
- simulation
- Task Types
- manipulation
- Episodes
- 100
- Data Format
- parquet
- Annotation Types
- reward_labelsaction_labels
- License
- mit
Access
Need custom proprioception data?
Claru builds purpose-built datasets for simulation applications with dense human annotations and quality assurance.
Request a Sample Pack