This repositoiry is built on the Flow software package to explore cyber-security attacks on intelligent transportation systems. Where Flow is focused on how sparesely adopted AVs can improve traffic, this repository explores the opposite question: How can sparsely cyber-compromised AVs be used to degrade traffic flow?
To recreate simulations from “Compromised ACC vehicles can degrade current mixed-autonomy traffic
performance while remaining stealthy against detection.” run \examples\full_network_attack.py to create attacked traffic.
For a guide on how to set up an adversarial simulation environment see \tutorials.
Follow these installation instructions to install Flow.
Cite the original flow repository using these papers:
C. Wu, A. Kreidieh, K. Parvate, E. Vinitsky, A. Bayen, “Flow: Architecture and Benchmarking for Reinforcement Learning in Traffic Control,” CoRR, vol. abs/1710.05465, 2017. [Online]. Available: https://arxiv.org/abs/1710.05465
Vinitsky, E., Kreidieh, A., Le Flem, L., Kheterpal, N., Jang, K., Wu, F., … & Bayen, A. M, Benchmarks for reinforcement learning in mixed-autonomy traffic. In Conference on Robot Learning (pp. 399-409). Available: http://proceedings.mlr.press/v87/vinitsky18a.html