A Methodology for the Development of RL-Based Adaptive Traffic Signal Controllers
This work addresses the problem of inconsistent comparisons in traffic signal control research for researchers and practitioners, but it is incremental as it focuses on methodology standardization rather than new algorithms.
The authors tackled the lack of standardization in developing adaptive traffic signal controllers using reinforcement learning by proposing a comprehensive methodology covering simulation setup, problem formulation, and experimental design, and they illustrated it in two simple scenarios to address limitations in current literature.
This article proposes a methodology for the development of adaptive traffic signal controllers using reinforcement learning. Our methodology addresses the lack of standardization in the literature that renders the comparison of approaches in different works meaningless, due to differences in metrics, environments, and even experimental design and methodology. The proposed methodology thus comprises all the steps necessary to develop, deploy and evaluate an adaptive traffic signal controller -- from simulation setup to problem formulation and experimental design. We illustrate the proposed methodology in two simple scenarios, highlighting how its different steps address limitations found in the current literature.