Solving Sum-of-Costs Multi-Agent Pathfinding with Answer-Set Programming

Gómez, Rodrigo; Hernandez, Carlos; Baier, Jorge

Keywords: artificial intelligence, MAPF

Abstract

Multi-Agent Pathfinding (MAPF) over grids is the problem of finding n non-conflicting paths that lead n agents from a given initial cell to a given goal cell. Cost-optimal MAPF in addition minimizes the total number of actions performed by each agent before stopping at the goal. Being a combinatorial problem in nature, a number of compilations from MAPF to Answer Set Programming (ASP) exist. In this paper we propose a new one, which unlike existing ASP approaches (1) produces cost-optimal solutions,(2) exploits information that can be pre-computed quickly using Dijkstra's algorithm, and (3) when grounded, produces a number of clauses that grows linearly with the number of agents. In our empirical evaluation, in which we use the clasp solver, we show that our approach is superior to heuristic-search-based algorithms in various settings.

Más información

Editorial: American Association for Artificial Intelligence (AAAI) Press
Fecha de publicación: 2020
Año de Inicio/Término: 7-12 Febrero
Página de inicio: 891
Página final: 897
Idioma: Inglés
URL: https://www.aaai.org/ocs/index.php/SOCS/SOCS19/paper/view/18374