Scalable Rail Planning and Replanning: Winning the 2020 Flatland Challenge

Multi-Agent Path Finding (MAPF) is the combinatorial problem of finding collision-free paths for multiple agents on a graph. This paper describes MAPF-based software for solving train planning and replanning problems on large-scale rail networks under uncertainty. The software recently won the 2020 Flatland Challenge, a NeurIPS competition trying to determine how to efficiently manage dense traffic on rail networks. The software incorporates many state-of-theart MAPF or, in general, optimization technologies, such as prioritized planning, large neighborhood search, safe interval path planning, minimum communication policies, parallel computing, and simulated annealing. It can plan collisionfree paths for thousands of trains within a few minutes and deliver deadlock-free actions in real-time during execution.

[1]  Howie Choset,et al.  PRIMAL: Pathfinding via Reinforcement and Imitation Multi-Agent Learning , 2018, IEEE Robotics and Automation Letters.

[2]  S. Mohanty,et al.  Flatland Competition 2020: MAPF and MARL for Efficient Train Coordination on a Grid World , 2021, NeurIPS.

[3]  Sven Koenig,et al.  Multi-Agent Path Finding with Delay Probabilities , 2016, AAAI.

[4]  Howie Choset,et al.  Path Planning for Multiple Agents under Uncertainty , 2017, ICAPS.

[5]  Roni Stern,et al.  Probabilistic Robust Multi-Agent Path Finding , 2020, SOCS.

[6]  Peter J. Stuckey,et al.  New Techniques for Pairwise Symmetry Breaking in Multi-Agent Path Finding , 2020, ICAPS.

[7]  Corina S. Pasareanu,et al.  Scheduling and Airport Taxiway Path Planning Under Uncertainty , 2019 .

[8]  Paul Shaw,et al.  Using Constraint Programming and Local Search Methods to Solve Vehicle Routing Problems , 1998, CP.

[9]  Peter J. Stuckey,et al.  Anytime Multi-Agent Path Finding via Large Neighborhood Search , 2021, AAMAS.

[10]  Erik Nygren,et al.  Flatland-RL : Multi-Agent Reinforcement Learning on Trains , 2020, ArXiv.

[11]  Sven Koenig,et al.  Multi-Agent Path Finding with Deadlines , 2018, IJCAI.

[12]  Jiaoyang Li,et al.  Moving Agents in Formation in Congested Environments , 2021, SOCS.

[13]  Gaurav S. Sukhatme,et al.  Trajectory Planning for Quadrotor Swarms , 2018, IEEE Transactions on Robotics.

[14]  Maxim Likhachev,et al.  SIPP: Safe interval path planning for dynamic environments , 2011, 2011 IEEE International Conference on Robotics and Automation.

[15]  Xiaolu Rao,et al.  Existing and novel Approaches to the Vehicle Rescheduling Problem (VRSP) , 2020 .

[16]  Keisuke Otaki,et al.  Multi-Agent Path Finding with Priority for Cooperative Automated Valet Parking , 2019, 2019 IEEE Intelligent Transportation Systems Conference (ITSC).

[17]  Jason M. O'Kane,et al.  Online Plan Repair in Multi-robot Coordination with Disturbances , 2019, 2019 International Conference on Robotics and Automation (ICRA).

[18]  Roni Stern,et al.  Multi-Agent Pathfinding: Definitions, Variants, and Benchmarks , 2019, SOCS.

[19]  Sven Koenig,et al.  Lifelong Multi-Agent Path Finding for Online Pickup and Delivery Tasks , 2017, AAMAS.

[20]  David Silver,et al.  Cooperative Pathfinding , 2005, AIIDE.

[21]  Nick Hawes,et al.  Multi-Robot Planning Under Uncertainty with Congestion-Aware Models , 2020, AAMAS.

[22]  Emilio Frazzoli,et al.  Provably safe and deadlock-free execution of multi-robot plans under delaying disturbances , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[23]  Akshat Kumar,et al.  Reinforcement Learning for Zone Based Multiagent Pathfinding under Uncertainty , 2020, ICAPS.

[24]  Andrew Tinka,et al.  Lifelong Multi-Agent Path Finding in Large-Scale Warehouses , 2020, AAMAS.

[25]  Nathan R. Sturtevant,et al.  Conflict-based search for optimal multi-agent pathfinding , 2012, Artif. Intell..

[26]  Sven Koenig,et al.  Optimal and Bounded-Suboptimal Multi-Agent Motion Planning , 2021, SOCS.

[27]  Marc Cavazza,et al.  Decentralized Multi-Agent Path Finding for UAV Traffic Management , 2019, IEEE Transactions on Intelligent Transportation Systems.

[28]  Roman Barták,et al.  Online Multi-Agent Pathfinding , 2019, AAAI.

[29]  David Pisinger,et al.  An Adaptive Large Neighborhood Search Heuristic for the Pickup and Delivery Problem with Time Windows , 2006, Transp. Sci..

[30]  Peter J. Stuckey,et al.  Searching with Consistent Prioritization for Multi-Agent Path Finding , 2018, AAAI.

[31]  Vadim Bulitko,et al.  Multi-Agent Pathfinding with Real-Time Heuristic Search , 2018, 2018 IEEE Conference on Computational Intelligence and Games (CIG).