disrupt_traffic

This repository contains the code needed to run the simulations detailed in our paper, How Well Do Reinforcement Learning Approaches Cope With Disruptions? The Case of Traffic Signal Control.

Setup

Installation can be done using

pip install -r requirements.txt

Running experiments

Training scripts may be found in disrupt_traffic/run_exp.py.

Citing

If you have found any part of this work useful or relevant to your research, please cite our article:

@article{korecki2023,
  author={Korecki, Marcin and Dailisan, Damian and Helbing, Dirk},
  journal={IEEE Access}, 
  title={How Well Do Reinforcement Learning Approaches Cope With Disruptions? The Case of Traffic Signal Control}, 
  year={2023},
  volume={11},
  number={},
  pages={36504-36515},
  doi={10.1109/ACCESS.2023.3266644}}

Acknowledgment

This work was funded by European Union’s Horizon 2020 Research and Innovation Programme through the Distributed Intelligence & Technology for Traffic & Mobility Management (DIT4TraM) (Grant Number: 953783)