Anytime Single-Step MAPF Planning with Anytime PIBT

2025-04-10

AI Summary

Prioritized Individual-Based Task (PIBT) achieves millisecond-scale response times but suffers from poor single-step solution quality due to its greedy priority assignment, and lacks any-time refinement—solution quality does not improve with extended planning time. Method: We propose Anytime PIBT, the first PIBT variant supporting anytime optimization. It integrates iterative priority reassignment, conflict-driven local rescheduling, and single-step optimality verification. Contribution/Results: We theoretically prove that Anytime PIBT converges to the single-step optimal solution. Empirically, on benchmarks with up to 100 agents, it significantly improves single-step solution quality within milliseconds and reaches theoretical optimality. Furthermore, we demonstrate that enhancing single-step optimality has limited impact on the global path cost, empirically validating the decoupling of single-step and global optimization—enabling efficient, scalable multi-agent path finding without sacrificing solution quality.

Technology Category

• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Heuristic Search

Application Category

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

Abstract

PIBT is a popular Multi-Agent Path Finding (MAPF) method at the core of many state-of-the-art MAPF methods including LaCAM, CS-PIBT, and WPPL. The main utility of PIBT is that it is a very fast and effective single-step MAPF solver and can return a collision-free single-step solution for hundreds of agents in less than a millisecond. However, the main drawback of PIBT is that it is extremely greedy in respect to its priorities and thus leads to poor solution quality. Additionally, PIBT cannot use all the planning time that might be available to it and returns the first solution it finds. We thus develop Anytime PIBT, which quickly finds a one-step solution identically to PIBT but then continuously improves the solution in an anytime manner. We prove that Anytime PIBT converges to the optimal solution given sufficient time. We experimentally validate that Anytime PIBT can rapidly improve single-step solution quality within milliseconds and even find the optimal single-step action. However, we interestingly find that improving the single-step solution quality does not have a significant effect on full-horizon solution costs.

Problem

• Improving greedy PIBT's poor solution quality in MAPF.
• Enabling continuous solution refinement within limited planning time.
• Exploring impact of single-step optimization on full-horizon costs.

Innovation

• Anytime PIBT improves solution quality continuously
• Converges to optimal solution with sufficient time
• Rapidly enhances single-step solutions within milliseconds