Solving Multi-Agent Pathfinding with Matching using A*+ID+OD
I.C. de Bruin (TU Delft - Electrical Engineering, Mathematics and Computer Science)
J. Mulderij – Mentor (TU Delft - Algorithmics)
Mathijs M. de Weerdt – Mentor (TU Delft - Algorithmics)
Marco A. Zuñiga Zamalloa – Graduation committee member (TU Delft - Embedded Systems)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
This paper extends the Multi-Agent Pathfinding (MAPF) algorithm, A*+ID+OD, to be able to solve problems with matching. This extension still keeps the optimal and completeness properties of the original algorithm. Matching is added to the algorithm in both an exhaustive and heuristic manner. Exhaustive matching is further improved by adding a new layer of Independence Detection (ID) to reduce the number of matchings. Besides this, the pruning efficiency is increased by sorting the matchings based on the initial heuristic. The exhaustive matching method has been found to perform better than the heuristic matching method. The exhaustive version of A*+ID+OD is finally compared to other extended MAPF algorithms which shows that on small maps, Conflict Based Min-Cost Flow (CBM) performs best as it is the only algorithm that does not use exhaustive matching. A*+ID+OD and Enhanced Partial Expansion A* (EPEA*) also perform well on open maps with multiple teams when compared to other exhaustive matching algorithms due to the addition of matching ID.