mulligan2026apache-2.0
real-marker-d2-c00-teleop-sobol
A teleoperator-controlled real robot manipulation dataset with 100 episodes collected using Sobol sampling variant. Part of the Mulligan robotics project with multiple wrist and side camera views.
Downloads23
Episodes100
Why This Matters for Physical AI
This dataset provides real-world teleoperated robot manipulation data with multi-view camera observations, supporting the development and evaluation of imitation learning policies for physical robotic systems.
Technical Profile
- Modalities
- rgb
- Environment
- lab
- Task Types
- manipulation
- Episodes
- 100
- Data Format
- parquet
- 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