Systematic Analyses of Reinforcement Learning Controllers in Signalized Urban Corridors
This work addresses traffic control optimization in urban corridors for improved efficiency, but it appears incremental as it builds on prior systematic analyses and focuses on specific network types.
The study extended systematic capacity region analysis to multi-junction traffic networks, specifically urban corridors, by training and evaluating centralized, decentralized, and parameter-sharing RL controllers against a MaxPressure baseline, showing that parameter-sharing controllers can generalize to larger networks and potentially enable self-organized 'green waves' without formal coordination.
In this work, we extend our systematic capacity region perspective to multi-junction traffic networks, focussing on the special case of an urban corridor network. In particular, we train and evaluate centralized, fully decentralized, and parameter-sharing decentralized RL controllers, and compare their capacity regions and ATTs together with a classical baseline MaxPressure controller. Further, we show how the parametersharing controller may be generalised to be deployed on a larger network than it was originally trained on. In this setting, we show some initial findings that suggest that even though the junctions are not formally coordinated, traffic may self organise into `green waves'.