Event
cuArms: A GPU-Accelerated Multi-Arm Conflict Planner
1 reports · 1 sources
Philip Huang and co-authors propose cuArms, a conflict-based multi-arm manipulation planner co-designed with GPU parallelism: it uses a reusable multimodal roadmap to decompose the search space and resolves conflicts and deadlocks through lazy search. The authors claim that, when connected to a high-level scheduler, the method can jointly optimize task allocation, scheduling, motion, and manipulation skills. The experimental conclusions reported by the authors are: cuArms can produce initial solutions in real time and deliver anytime improvements in motion quality and scheduling, without requiring the multi-day domain training that learning-based methods need; the authors state that demonstration videos of both simulation and real industrial tasks have been made public. These results are the authors' own claims, and the report provides no independent verification.
Report timeline
- cuArms: A GPU-Accelerated Conflict-Based Planner for Multi-Arm Manipulation
cuArms is a conflict-based planner co-designed with GPU parallelism. It decomposes the search space using reusable multi-modal roadmaps and resolves conflicts and deadlocks through lazy search. It integrates with a high-level scheduler to jointly optimize task assignment, scheduling, motion, and manipulation skills. Experiments report real-time initial solutions and anytime improvements in motion quality and scheduling, without domain training for learning-based methods. The authors say it provides demo videos on both simulated and real industrial tasks.
arXiv cs.RO ·