Reevaluation of Large Neighborhood Search for MAPF: Findings and Opportunities

TL;DR

This paper reevaluates large neighborhood search methods for multi-agent pathfinding, highlighting the strengths of rule-based heuristics, exposing limitations of current learning-based approaches, and proposing a unified evaluation framework for future research.

Contribution

It introduces a unified evaluation framework for MAPF-LNS, compares existing methods comprehensively, and identifies new research opportunities for integrating machine learning.

Findings

Rule-based heuristics outperform many learning-based methods.

Current learning methods show no clear advantage in efficiency or improvement.

A unified evaluation framework reveals gaps and guides future MAPF-LNS research.

Abstract

Multi-Agent Path Finding (MAPF) aims to arrange collision-free goal-reaching paths for a group of agents. Anytime MAPF solvers based on large neighborhood search (LNS) have gained prominence recently due to their flexibility and scalability, leading to a surge of methods, especially those leveraging machine learning, to enhance neighborhood selection. However, several pitfalls exist and hinder a comprehensive evaluation of these new methods, which mainly include: 1) Lower than actual or incorrect baseline performance; 2) Lack of a unified evaluation setting and criterion; 3) Lack of a codebase or executable model for supervised learning methods. To address these challenges, we introduce a unified evaluation framework, implement prior methods, and conduct an extensive comparison of prominent methods. Our evaluation reveals that rule-based heuristics serve as strong baselines, while current learning-based methods show no clear advantage on time efficiency or improvement capacity. Our extensive analysis also opens up new research opportunities for improving MAPF-LNS, such as targeting high-delayed agents, applying contextual algorithms, optimizing replan order and neighborhood size, where machine learning can potentially be integrated. Code and data are available at https://github.com/ChristinaTan0704/mapf-lns-unified.