Event
AuRA: Autonomous Robotic Interaction for Digital Twin Reconstruction
1 reports · 1 sources
Xinchen Yu and co-authors propose AuRA, an automated robotic interaction and data preparation system for digital twin reconstruction that delegates the manual collection and preparation pipeline to robots. The system combines execution-aware grasp selection, gripper motion measurement derived from robot state, and registered pre-interaction geometry and appearance priors. On 13 deformable objects, the authors report reductions of 88.5% in human effort and 55.4% in actual collection time per successfully reconstructed asset. Compared with the manually collected PhysTwin baseline, in future prediction under known gripper motion, 7 of 9 metrics improve, with one-way L1 Chamfer distance reduced by 28.1%; on the evaluated unseen interactions, mean contour IoU and Chamfer distance are similar to the baseline.
Report timeline
- AuRA: Reconstructing Digital Twins from Atoms to Bits through Autonomous Robot Interaction
AuRA is an automated robot interaction and data preparation system for digital-twin reconstruction, combining execution-aware grasp selection, gripper-motion measurements derived from robot states, and registered pre-interaction geometry and appearance priors. Across 13 deformable objects, it reduces active human labor and wall-clock collection time per successfully reconstructed asset by 88.5% and 55.4%, respectively; compared with the human-collected PhysTwin baseline, it improves 7 of 9 metrics for future prediction under known gripper motion, including a 28.1% reduction in one-way L1 Chamfer distance.
arXiv cs.RO ·