Flow is a deep reinforcement learning framework for mixed autonomy traffic. Flow is a traffic control benchmarking framework and it provides a suite of traffic control scenarios (benchmarks), tools for designing custom traffic scenarios, and integration with deep reinforcement learning and traffic microsimulation libraries.

Yes! Please refer to our Installation instructions.

Yes! In fact Flow has rich documentaion and we frequently update them. Please refer to Flow documentation.

Yes! Flow is open-source for public use, and it is licensed under the MIT license.

You can report bugs by submitting GitHub issues. To submit a github issue, please click here

Yes! Please check our Tutorial Page to get started with deep reinforcement learning and transportation. We also have Python Jupyter Tutorials for Flow.

Thank you for your interest in contributing to the project Flow. Please Submit your contributions using Github pull requests. Click here to submit a new pull request.

If you use Flow for academic research, you are highly encouraged to cite this paper:

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. Available: Arxive

If you use the benchmarks, you are highly encouraged to cite this paper:

E. Vinitsky, A. Kreidieh, L. L. Flem, N. Kheterpal, K. Jang, C. Wu, F. Wu, R. Liaw, E. Liang, A. M. Bayen. "Benchmarks for reinforcement learning in mixed-autonomy traffic". In Conference on Robot Learning (pp. 399-409). Available: PMLR