ajaysri
Three-holder cable routing with vector goals
Real-robot demonstrations for a vector-conditioned low-level policy where an ARX bimanual robot places three holders and routes a cable through each holder. The dataset contains 840 subtask episodes from 140 source recordings with multi-view RGB observations and vector goal annotations.
Downloads0
Episodes840
Why This Matters for Physical AI
This dataset provides real-robot manipulation demonstrations with vector-based goal specifications, enabling training of vision-conditioned policies for complex multi-step cable manipulation tasks that require precise spatial reasoning and object interaction.
Technical Profile
- Modalities
- rgbproprioceptionlanguage
- Robot Embodiments
- ARX bimanual
- Action Space
- joint_positions
- Environment
- lab
- Task Types
- manipulationcable_routingplacement
- Episodes
- 840
- Data Format
- LeRobot
- Annotation Types
- language_instructionsvector_goalsaction_labels
Access
Need custom rgb data?
Claru builds purpose-built datasets for lab applications with dense human annotations and quality assurance.
Request a Sample Pack