mulligan2026MIT
real-marker-d2-c00-teleop-mixed
A real robot manipulation dataset collected via teleoperation containing 250 episodes of marker-based task demonstrations. Part of the Mulligan project, providing raw teleoperated collection data with multi-camera observations.
Downloads330
Episodes250
Hours0.8882
Why This Matters for Physical AI
This dataset contributes to physical AI research by providing real robot teleoperated demonstrations with multi-camera views that can be used for imitation learning and policy training in manipulation tasks.
Technical Profile
- Modalities
- rgb
- Robot Embodiments
- franka
- Action Space
- 7-DoF: vel_x, vel_y, vel_z, vel_roll, vel_pitch, vel_yaw, gripper_action
- Environment
- lab
- Task Types
- manipulation
- Episodes
- 250
- Total Hours
- 0.8882
- 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