Multi-Agent Path Finding

Also known as MAPF

Multi-agent path finding studies how to route agents to their goals without collisions, often minimizing travel cost or completion time; in lifelong settings, agents repeatedly receive new tasks. Classical methods include Conflict-Based Search and space-time search for high-quality or optimal plans, while prioritized planning, PIBT, and LaCAM trade some solution quality for scalability. Other approaches use learning, local communication, optimization, or hybrid methods that combine fast planning with conflict repair, and many now address task assignment, congestion, and execution on real robots. A central tension is balancing guarantees and realistic motion or disturbances against the speed and scale needed for dense fleets.

Latest papers

93 papers13 papers in the last four weeks, up 44% on the four weeks before. 0.1% of all new papers.

MAPF-World: Action World Model for Multi-Agent Path Finding
Aug 16, 2025Zhanjiang Yang, Yueming Li, Yang Shen +2Multi-Agent Path Finding Multi-Agent Planning
Shenzhen Technology University, China. · University of Technology Sydney, Australia.
Simultaneous Computation with Multiple Prioritizations in Multi-Agent Motion Planning
Jan 18, 2025Patrick Scheffe, Julius Kahle, Bassam AlrifaeeDecentralized Multi-Agent Motion Planning Motion Planning
Chair of Embedded Software, RWTH Aachen University, Im Süsterfeld 9, 52072 Aachen, Germany · Department of Aerospace Engineering, University of the Bundeswehr Munich, Werner-Heisenberg-Weg 39, 85579 Neubiberg, Germany
Multi-Agent Reinforcement Learning-Based UAV Pathfinding for Obstacle Avoidance in Stochastic Environment
Oct 25, 2023Qizhen Wu, Kexin Liu, Lei Chen +1Mean-Field MARL Multi-Agent Path Finding
School of Automation Science and Electrical Engineering, Beihang University, Beijing 100191, China. · Advanced Research Institute of Multidisciplinary Sciences and State Key Laboratory of CNS/ATM, Beijing Institute of Technology, Beijing 100081, China.

Aug 16, 2025J/K move · Enter open · S save

MAPF-World: Action World Model for Multi-Agent Path Finding

Zhanjiang Yang, Yueming Li, Yang Shen, Meng Li+1

Shenzhen Technology University, China. · University of Technology Sydney, Australia.

Multi-Agent Path Finding Multi-Agent Planning Multi-Agent Coordination World Model-Based Planning World Models

Open paper AskPDF arXiv

Multi-agent path finding (MAPF) studies the problem of planning conflict-free paths for multiple agents from given start locations to designated goals, with applications in robot-assisted logistics and social navigation. Recent decentralized learned solvers have shown promise for large-scale MAPF, particularly when leveraging foundation models and large datasets. However, most existing methods rely on reactive policies, often resulting in congestion, deadlocks, and degraded generalization in high agent-density environments. To address these limitations, we propose MAPF-World, an autoregressive action world model for MAPF that unifies short-horizon local future prediction and action generation, enabling decision-making beyond immediate local observations. MAPF-World models short-horizon local dynamics by predicting the next local observation and neighboring agents' action intentions, capturing both spatial structures and temporal interaction patterns. We further introduce a spatio-agent positional encoding that integrates spatial awareness with agent-level semantics in Transformer-based architectures, facilitating more coordinated multi-agent behaviors. In addition, we augment existing MAPF benchmarks by introducing an automated map generator grounded in real-world urban layouts, aiming to narrow the gap between synthetic simulation and practical deployment scenarios. Extensive experiments across diverse map types and interaction settings demonstrate that MAPF-World achieves strong performance compared with existing learned solvers. Notably, it exhibits robust zero-shot generalization and maintains a high success rate even as agent density increases.

Latest papers

MAPF-World: Action World Model for Multi-Agent Path Finding

Simultaneous Computation with Multiple Prioritizations in Multi-Agent Motion Planning

Multi-Agent Reinforcement Learning-Based UAV Pathfinding for Obstacle Avoidance in Stochastic Environment