论文信息 - Don't Split, Try To Work It Out: Bypassing Conflicts in Multi-Agent Pathfinding

Don't Split, Try To Work It Out: Bypassing Conflicts in Multi-Agent Pathfinding

Conflict-Based Search (CBS) is a recently introduced algorithm for Multi-Agent Path Finding (MAPF) whose runtime is exponential in the number of conflicts found between the agents' paths. We present an improved version of CBS that bypasses conflicts thereby reducing the CBS search tree. Experimental results show that this improvement reduces the runtime by an order of magnitude in many cases.

[1] Roni Stern,et al. The Increasing Cost Tree Search for Optimal Multi-Agent Pathfinding , 2011, IJCAI.

[2] Nathan R. Sturtevant,et al. Meta-Agent Conflict-Based Search For Optimal Multi-Agent Path Finding. , 2012 .

[3] Nathan R. Sturtevant,et al. Conflict-based search for optimal multi-agent pathfinding , 2012, Artif. Intell..

[4] Pavel Surynek,et al. Towards Optimal Cooperative Path Planning in Hard Setups through Satisfiability Solving , 2012, PRICAI.

[5] Peter Schüller,et al. A General Formal Framework for Pathfinding Problems with Multiple Agents , 2013, AAAI.

[6] Trevor Scott Standley. Finding Optimal Solutions to Cooperative Pathfinding Problems , 2010, AAAI.

[7] Nathan R. Sturtevant,et al. Benchmarks for Grid-Based Pathfinding , 2012, IEEE Transactions on Computational Intelligence and AI in Games.

[8] Steven M. LaValle,et al. Planning optimal paths for multiple robots on graphs , 2012, 2013 IEEE International Conference on Robotics and Automation.

[9] Roni Stern,et al. Using Lookaheads with Optimal Best-First Search , 2010, AAAI.

[10] Nathan R. Sturtevant,et al. Partial-Expansion A* with Selective Node Generation , 2012, SOCS.

[11] Howie Choset,et al. M*: A complete multirobot path planning algorithm with performance bounds , 2011, 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems.