Multimodal Multiscale Signal-Vehicle Coupled Control

New advancements enabled by the connected and automated vehicles (CAVs) has made possible the effective cooperation between road users and traffic infrastructure towards more sustainable, efficient and safe mobility in transportation networks. This study extends a Signal-Vehicle Coupled Control (SVCC) model to optimize traffic signals at intersections for various transportation modes, including active transportation (Pedestrians and Cyclists) and vehicles with different power sources (e.g.,fuel, hybrid and electric) and CAVs. An optimization framework is developed using a Model Predictive Control (MPC) scheme, capable of coordinating CAVs and signal controllers under various infrastructure configurations. These include concurrent and exclusive pedestrian and bicycle phasing, various vehicle turning treatments, as well as mixed and separated bike flows. The results demonstrate that integrating pedestrians, cyclists, and vehicle heterogeneity into the optimization process through
the proposed Multimodal Multiscale SVCC (M2SVCC) yields superior performance compared to conventional actuated signal plans. Specifically, M2 SVCC achieves significant reductions in fuel and energy consumption (sustainability), user delays (mobility), and traffic conflicts (safety) across a range of active user demands and signal configurations.