Local Guidance for Configuration-Based Multi-Agent Pathfinding

2025-10-21

AI Summary

To address low global coordination efficiency, local congestion, and prolonged waiting times in configuration-based multi-agent path finding (MAPF), this paper proposes a local guidance approach. Within the LaCAM framework, it dynamically decomposes global guidance information into spatiotemporal constraints localized to each agent’s neighborhood, enabling real-time re-planning. By bypassing computationally expensive global optimization, the method achieves a favorable trade-off between solution quality and runtime efficiency. Experiments demonstrate that, under reasonable time budgets, our approach significantly reduces average waiting time—by up to 42%—while improving path success rate and system throughput, and maintaining strong robustness in dense, dynamic environments. The core contribution is the first scalable transformation of global guidance into lightweight, adaptive local directives—a breakthrough that advances the performance frontier of configuration-based MAPF solvers.

Technology Category

• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: Localization, Mapping, and Navigation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Guidance is an emerging concept that improves the empirical performance of real-time, sub-optimal multi-agent pathfinding (MAPF) methods. It offers additional information to MAPF algorithms to mitigate congestion on a global scale by considering the collective behavior of all agents across the entire workspace. This global perspective helps reduce agents' waiting times, thereby improving overall coordination efficiency. In contrast, this study explores an alternative approach: providing local guidance in the vicinity of each agent. While such localized methods involve recomputation as agents move and may appear computationally demanding, we empirically demonstrate that supplying informative spatiotemporal cues to the planner can significantly improve solution quality without exceeding a moderate time budget. When applied to LaCAM, a leading configuration-based solver, this form of guidance establishes a new performance frontier for MAPF.

Problem

• Enhancing multi-agent pathfinding with local guidance
• Reducing agent congestion through spatiotemporal cues
• Improving solution quality within moderate computational limits

Innovation

• Local guidance reduces agent congestion
• Spatiotemporal cues improve solution quality
• Applied to LaCAM solver for MAPF