Enhancing Lifelong Multi-Agent Path-finding by Using Artificial Potential Fields

TL;DR

This paper investigates the application of Artificial Potential Fields to improve Lifelong Multi-Agent Path Finding, demonstrating significant throughput gains in lifelong scenarios but not in standard MAPF.

Contribution

It introduces methods for integrating APFs into existing MAPF algorithms and evaluates their effectiveness in lifelong pathfinding tasks.

Findings

APFs do not improve standard MAPF performance.

APFs increase system throughput up to 7 times in Lifelong MAPF.

The proposed methods enhance efficiency in continuous goal scenarios.

Abstract

We explore the use of Artificial Potential Fields (APFs) to solve Multi-Agent Path Finding (MAPF) and Lifelong MAPF (LMAPF) problems. In MAPF, a team of agents must move to their goal locations without collisions, whereas in LMAPF, new goals are generated upon arrival. We propose methods for incorporating APFs in a range of MAPF algorithms, including Prioritized Planning, MAPF-LNS2, and Priority Inheritance with Backtracking (PIBT). Experimental results show that using APF is not beneficial for MAPF but yields up to a 7-fold increase in overall system throughput for LMAPF.