Reinforcement Learning Benchmarks for Traffic Signal Control Academic Article uri icon

abstract

  • We propose a toolkit for developing and comparing reinforcement learning (RL)-based traffic signal controllers. The toolkit includes implementation of state-of-the-art deep-RL algorithms for signal control along with benchmark control problems that are based on realistic traffic scenarios. Importantly, the toolkit allows a first-of-its-kind comparison between state-of-the-art RL-based signal controllers while providing benchmarks for future comparisons. Consequently, we compare and report the relative performance of current RL algorithms. The experimental results suggest that previous algorithms are not robust to varying sensing assumptions and non-stylized intersection layouts. When more realistic signal layouts and advanced sensing capabilities are assumed, a distributed deep-Q learning approach is shown to outperform previously reported state-of-the-art algorithms in many cases.

published proceedings

  • the 35th Neural Information Processing Systems (NeurIPS 2021) Track on Datasets and Benchmarks

author list (cited authors)

  • Ault, J., & Sharon, G.

complete list of authors

  • Ault, James||Sharon, Guni

publication date

  • December 2021