Adaptive Traffic Light Control With Quality-of-Service Provisioning for Connected and Automated Vehicles at Isolated Intersections

Today’s traffic intersections primarily use fixed-cycle signaling, which is highly suboptimal and aggravates congestion and waste of fuel in urban areas. In addition, it offers minimal quality-of-service guarantee and makes travel time prediction hard. In this paper, we study how traffic signal scheduling can be done in order to not only reduce average wait time but also provide certain worst-case wait time guarantee to Connected and Automated Vehicles (CAVs). First, we derive the time a CAV spends at an intersection in the best-case and the worst-case, respectively. Using the worst-case wait time result, we propose two novel controllers which assign deadlines to CAVs based on the system state and then perform traffic signal control to meet these deadlines. We use the Simulation of Urban MObility (SUMO) simulator to implement and verify the performance of the proposed controllers. The simulation results show that they not only can guarantee the deadlines, but also outperform other methods such as the default and the delay-based controllers in terms of average delay in all scenarios.