Robust MAPF

MAPF: Multi-Agent Path Finding

Robust MAPF studies how multiple robots can reach their goals without collisions when plans face delays, uncertain actions or observations, and changing tasks. Methods include prioritized planning, conflict-based search, temporal flexibility and replanning, deadlock prevention, and learning-based policies, often combined with rolling-horizon planning for lifelong warehouse and industrial fleets. Evaluations range from grid benchmarks to dense warehouse simulations and physical-robot settings, with measures such as success rate, throughput, runtime, and delay impact. A central challenge is balancing reliable execution and solution guarantees against the speed and scale needed to coordinate large fleets in congested environments.

Latest papers

18 papers1 paper in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Reinforcement Learning for Large-Scale Lifelong Multi-Agent Path Finding with Shared Spatial Fields
Oct 13, 2026Jean Seong Bjorn ChoeMulti-Agent Path Finding Multi-Agent Planning
Independent Researcher
A traffic management system for large and heterogeneous vehicles in narrow industrial environments
Sep 9, 2026Alessandro Bonetti, Silvia Proia, Simone Guidetti +1Robot Navigation Robust MAPF
Department of Sciences and Methods for Engineering (DISMI), University of Modena and Reggio Emilia, Italy · Gruppo TecnoFerrari S.p.a. con socio unico, Italy
AOC-CBS: Anytime-Optimal Continuous-time Conflict-Based Search for Generalised Multi-Agent Path Finding
Aug 8, 2026Alvin Combrink, Sabino Francesco Roselli, Martin FabianRobust MAPF Conflict-Based Search
Department of Electrical Engineering, Chalmers University of Technology, Gothenburg, Sweden.
Complete, Scalable, and Robust Prioritized Planning for Multi-Robot Ordered Storage and Retrieval at Maximum Capacity
Aug 7, 2026William Zhang, Tzvika Geft, Jingjin Yu +1Multi-Robot Motion Planning Multi-Robot Systems
Department of Computer Science, Rutgers University, New Brunswick, NJ, USA
Scalable Long-Horizon Planning with Staggered Updates for Lifelong MAPF
Aug 7, 2026Vaibhav Sanjay, Jiaoyang LiMulti-Agent Coordination Robust MAPF
Robotics Institute, Carnegie Mellon University
Search-Aided Joint Agent-Environment Reinforcement Learning for Robust Lifelong Multi-Agent Path Finding with Rotations
Aug 6, 2026He Jiang, Jingtian Yan, Yulun Zhang +5Constrained Motion Planning Robust MAPF
1Carnegie Mellon University · University of Southern California · 3National University of Singapore
Cooperative-ORCA*: Real-Time Proactive Deadlock Avoidance for Continuous-Space Multi-Agent Navigation
Jun 22, 2026Junfeng Wu, Jiaqi Chen, Hongkun Lyu +2Multi-Robot Motion Planning Robust MAPF
Monash University
Intermittent Strategic Cooperation of Two Selfish Agents on Graphs
Jun 15, 2026Itay Shedlezki, Noa AgmonNash Equilibrium Game Theory
Bar-Ilan University · Ramat Gan, Israel
Lifelong LaCAM with Local Guidance for Lifelong MAPF
May 16, 2026Tomoki Arita, Keisuke OkumuraSequential Decision Making Robust MAPF
Keio University, Japan · National Institute of Advanced Industrial Science and Technology (AIST), Japan
Conveyor Parcel Routing with Order-Contiguous Arrivals
May 13, 2026Takuro Kato, Keisuke OkumuraRobust MAPF Multi-Agent Path Finding
Toyota Industries Corporation, Japan · National Institute of Advanced Industrial Science and Technology (AIST), Japan
Robust Multi-Agent Path Finding under Observation Attacks: A Principled Adversarial-Plus-Smoothing Training Recipe
May 12, 2026Riad AhmedMulti-Agent Path Finding Adversarial Robustness
University of New Hampshire
Learning to Communicate Locally for Large-Scale Multi-Agent Pathfinding
May 8, 2026Valeriy Vyaltsev, Alsu Sagirova, Anton Andreychuk +5Multi-Agent Coordination Robust MAPF
CogAI Lab, Moscow, Russia
Should I Replan? Learning to Spot the Right Time in Robust MAPF Execution
Apr 28, 2026David Zahrádka, David Woller, Denisa Mužíková +2Robust MAPF Multi-Agent Path Finding
Czech Institute of Informatics, Robotics and Cybernetics, Czech Technical University in Prague · Faculty of Electrical Engineering, Czech Technical University in Prague · Czech Institute of Informatics, Robotics and Cybernetics, Czech Technical University in Prague, Czechia. +1
CTS-PLL: A Robust and Anytime Framework for Collaborative Task Sequencing and Multi-Agent Path Finding
Mar 26, 2026Junkai Jiang, Yitao Xu, Ruochen Li +2Multi-Agent Planning Multi-Agent Path Finding
Learning-guided Prioritized Planning for Lifelong Multi-Agent Path Finding in Warehouse Automation
Mar 25, 2026Han Zheng, Yining Ma, Brandon Araki +2Warehouse Automation Multi-Agent Path Finding
Massachusetts Institute of Technology, Cambridge, MA, USA · Symbotic, Wilmington, MA, USA
Route Fragmentation Based on Resource-centric Prioritisation for Efficient Multi-Robot Path Planning in Agricultural Environments
Mar 13, 2026James R. Heselden, Gautham P. DasMulti-Robot Motion Planning Robust MAPF
Lincoln Centre for Autonomous Systems, University of Lincoln, UK. · Lincoln Institute for Agri-food Technology, University of Lincoln, UK.
Lifelong Scalable Multi-Agent Realistic Testbed and A Comprehensive Study on Design Choices in Lifelong AGV Fleet Management Systems
Feb 17, 2026Jingtian Yan, Yulun Zhang, Zhenting Liu +5Multi-Robot Systems Multi-Agent Path Finding
Robotics Institute, Carnegie Mellon University · Symbotic
Precomputing Multi-Agent Path Replanning Using Temporal Flexibility
Jan 8, 2026Issa Hanou, Eric Kemmeren, Devin Wild Thomas +1Safe Interval Path Planning Multi-Agent Path Finding
Delft University of Technology · University of New Hampshire

Oct 13, 2026J/K move · Enter open · S save

Reinforcement Learning for Large-Scale Lifelong Multi-Agent Path Finding with Shared Spatial Fields

Jean Seong Bjorn Choe

Independent Researcher

Multi-Agent Path Finding Multi-Agent Planning Multi-Agent Reinforcement Learning Robust MAPF

Open paper AskPDF arXiv

Large-scale applications of lifelong multi-agent path finding (LMAPF) call for computationally inexpensive movement decisions under changing traffic conditions and across varied layouts. We propose a navigation policy in which a convolutional network computes a shared spatial field over the map once per timestep and a compact goal-conditioned operator derives agents' movement preferences. The complete policy is trained end-to-end using average-reward reinforcement learning with throughput as the reward, without planner demonstrations or a separate online path planner. For structured and unstructured layouts, we train one policy per class on mixtures of small maps and deploy each without retraining on larger, unseen layouts and agent populations. Across six layouts with 10000 agents, our policies achieve throughput comparable to or higher than the strongest evaluated baselines. The proposed method has an average execution time below 10 ms per step, more than an order of magnitude faster than several established planners.