A curated list of awesome papers, websites, videos, and repositories related to Multi Agent Path Finding (MAPF).
Welcome to the MAPF resources repository, an organized collection of valuable materials related to Multi Agent Path Finding (MAPF).
MAPF is the problem of computing collision-free paths for a team of agents from their current locations to given destinations. (mapf.info)
Our primary objective is to build and maintain a classified, community-driven assortment of widely recognized resources. This collection can serve as a reference point, a springboard for research, or simply as a well-curated list of resources for those interested in MAPF.
Please note, the code shared here may not always be the official version associated with the original paper. Therefore, we strongly advise caution and thorough examination when utilizing the code for your purposes.
Your questions, resource requests, and pull request contributions are always welcome and appreciated. Together, we can create a comprehensive and useful hub for MAPF resources!
- Survey
- Multi-Agent Path Finding – An Overview
- Multi-Agent Pathfinding: Definitions, Variants, and Benchmarks
- Problem Compilation for Multi-Agent Path Finding: a Survey
- Planning and Learning in Multi-Agent Path Finding
- Structure and Intractability of Optimal Multi-Robot Path Planning on Graphs (NP-hardness proof)
- Search-based Approach
- Cooperative Pathfinding [1] [2] (Code)
- Conflict-based search for optimal multi-agent pathfinding (CBS) (Code)
- Suboptimal Variants of the Conflict-Based Search Algorithm for the Multi-Agent Pathfinding Problem (ECBS) (Code)
- EECBS: A Bounded-Suboptimal Search for Multi-Agent Path Finding (EECBS) (Code)
- ICBS: Improved Conflict-Based Search Algorithm for Multi-Agent Pathfinding (ICBS) (Code)
- Multi-Agent Pathfinding with Continuous Time (CCBS) (Code)
- Subdimensional expansion for multirobot path planning (M*) (Code)
- SIPP: Safe Interval Path Planning for Dynamic Environments (SIPP) (Code)
- Anytime Multi-Agent Path Finding via Large Neighborhood Search (MAPF-LNS) (Code)
- MAPF-LNS2: Fast Repairing for Multi-Agent Path Finding via Large Neighborhood Search (MAPF-LNS2) (Code)
- Sampling-based Approach
- Learning-based Approach
- PRIMAL: Pathfinding via Reinforcement and Imitation Multi-Agent Learning (PRIMAL) (Code)
- PRIMAL2: Pathfinding via Reinforcement and Imitation Multi-Agent Learning - Lifelong (PRIMAL2) (Code)
- CTRMs: Learning to Construct Cooperative Timed Roadmaps for Multi-agent Path Planning in Continuous Spaces (CTRM) (Project) (Code)
- mapf.info
- MAPF Visualizer
- MAPF Benchmark
- Foundations of Multi-Agent Path Finding
- Project "Multi-Agent Path Planning (MAPF)"
- AAAI-23 Workshop on Multi-Agent Path Finding
- AAMAS-22 Tutorial on Recent Advances in Multi-Agent Path Finding