Standardized feature extraction from pairwise conflicts applied to the train rescheduling problem

2022/04/06 by Anikó Kopacz, Kopacz, Anikó, Ágnes Mester +5
Engineering · #Artificial Intelligence (cs.AI) #Elevator Systems and Control #FOS: Computer and information sciences #Machine Learning (cs.LG) #Railway Systems and Energy Efficiency

paper · pdf · doi:10.48550/arxiv.2204.03061

openalex publication_date 2022/04/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We propose a train rescheduling algorithm which applies a standardized feature selection based on pairwise conflicts in order to serve as input for the reinforcement learning framework. We implement an analytical method which identifies and optimally solves every conflict arising between two trains, then we design a corresponding observation space which features the most relevant information considering these conflicts. The data obtained this way then translates to actions in the context of the reinforcement learning framework. We test our preliminary model using the evaluation metrics of the Flatland Challenge. The empirical results indicate that the suggested feature space provides meaningful observations, from which a sensible scheduling policy can be learned.

Related

• Improving Sample Efficiency and Multi-Agent Communication in RL-based Train Rescheduling
• Flatland-RL : Multi-Agent Reinforcement Learning on Trains
• Reinforcement Learning for Scalable Train Timetable Rescheduling with Graph Representation
• Railway Operation Rescheduling System via Dynamic Simulation and Reinforcement Learning
• Flatland Competition 2020: MAPF and MARL for Efficient Train Coordination on a Grid World