Reinforcement learning (RL) is gaining popularity as an effective approach for traffic signal control (TSC) and is increasingly applied in this domain. However, most existing RL methodologies are confined to a single-stage TSC framework, primarily focusing on selecting an appropriate traffic signal phase at fixed action intervals, leading to inflexible and less adaptable phase durations. To address such limitations, we introduce a novel two-stage TSC framework named DynamicLight. This framework initiates with a phase control strategy responsible for determining the optimal traffic phase, followed by a duration control strategy tasked with determining the corresponding phase duration. Experimental results show that DynamicLight outperforms state-of-the-art TSC models and exhibits exceptional model generalization capabilities. Additionally, the robustness and potential for real-world implementation of DynamicLight are further demonstrated and validated through various DynamicLight variants. The code is released at https://github.com/LiangZhang1996/DynamicLight.
@article{arxiv.2211.01025,
title = {DynamicLight: Two-Stage Dynamic Traffic Signal Timing},
author = {Liang Zhang and Yutong Zhang and Shubin Xie and Jianming Deng and Chen Li},
journal= {arXiv preprint arXiv:2211.01025},
year = {2024}
}