Multi-Agent Corridor Generating Algorithm

2024-10-16

AI Summary

In dense multi-agent pathfinding (MAPF) scenarios, existing methods suffer from frequent collisions, deadlocks, and degraded performance. To address this, we propose a cooperative distributed pathfinding framework based on dynamic corridor construction. Each agent incrementally generates a connected vertex sequence—termed a “corridor”—toward its goal and actively coordinates with nearby agents to resolve conflicts via corridor-guided evacuation. We introduce the first corridor-driven distributed collision avoidance mechanism and integrate the rule-based PIBT algorithm into the corridor generation framework, yielding the hybrid MACGA+PIBT algorithm. This design ensures polynomial-time complexity, global reachability, theoretical solvability, and practical efficiency. Evaluated on standard MAPF grid benchmarks, our method significantly outperforms state-of-the-art baselines in success rate, runtime, and makespan.

Technology Category

• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: Motion and Path Planning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

In this paper, we propose the Multi-Agent Corridor Generating Algorithm (MACGA) for solving the Multi-agent Pathfinding (MAPF) problem, where a group of agents need to find non-colliding paths to their target locations. Existing approaches struggle to solve dense MAPF instances. In MACGA, the agents build emph{corridors}, which are sequences of connected vertices, from current locations towards agents' goals, and evacuate other agents out of the corridors to avoid collisions and deadlocks. We also present the MACGA+PIBT algorithm, which integrates the well-known rule-based PIBT algorithm into MACGA to improve runtime and solution quality. The proposed algorithms run in polynomial time and have a reachability property, i.e., every agent is guaranteed to reach its goal location at some point. We demonstrate experimentally that MACGA and MACGA+PIBT outperform baseline algorithms in terms of success rate, runtime, and makespan across diverse MAPF benchmark grids.

Problem

• Solves Multi-agent Pathfinding (MAPF) with non-colliding paths
• Addresses dense MAPF instances using corridor generation
• Ensures reachability and improves runtime with MACGA+PIBT

Innovation

• Multi-agent corridors for collision avoidance
• Integration with PIBT for efficiency
• Polynomial-time reachability guarantees