Event
Revision of the CARE Human-Guided Collaborative Carrying Method Paper
1 reports · 1 sources
Ning Liu and other authors have released a revised version of the paper on the CARE human-guided robot collaborative carrying method. CARE grounds local control in "human-robot joint transitions indexed by executable actions," allowing the same executable action to simultaneously determine operator response prediction, human-robot joint rollout, and candidate selection. The authors report that on training and held-out maps, CARE maintains near-saturated performance when the instructed route remains locally traversable; when the operator can continue but the robot must detour, CARE achieves the highest success rate and service availability. The revision also adds a bounded reciprocity correction to handle brief mismatches between route and feasibility; when progress is persistently lost, it switches to a progress-oriented recovery proposal, while downstream still uses the same executable-action-conditioned evaluation. The authors report that an indoor deployment demonstrated the method's physical executability.
Report timeline
- CARE: Consistent Action-Response Execution for Human-Guided Robot Transport
The authors propose Consistent Action-Response Execution (CARE), which grounds local control in executable-action-indexed joint human-robot transitions, so that the same executable action simultaneously determines operator response prediction, joint human-robot rollout, and candidate selection. On training and held-out maps, CARE maintains near-saturated performance when the indicated route remains locally passable, and achieves the highest success rate and service availability when the operator can continue but the robot must detour.
arXiv cs.RO ·