Integrated Optimization of Vehicle Microscopic Behavior and Signal Control for Mixed Traffic Based on a Distributed Strategy

This paper proposes a bilevel optimization framework to coordinate intersection signal control and vehicle behavior in mixed traffic. The upper-level optimization centrally manages adaptive signal timing, adjusting green phase durations and introducing a white phase that allows selected vehicles to pass through the intersection during red signals to reduce idle time. The lower-level optimization provides decentralized vehicle-level control for connected and automated vehicles, optimizing their trajectories and microscopic driving behaviors as they interact with human-driven vehicles. The framework takes advantage of both signal control and conflict elimination by incorporating an integrated level and a decentralized level. Simulation results demonstrated that the proposed approach improved traffic efficiency and stability while reducing energy consumption compared with conventional signal control methods.