Reservation-Prioritization-Based Mixed-Traffic Cooperative Control at Unsignalized Intersections

Connected and Autonomous Vehicle (CAV) has attracted much attention as it provides promising solutions to improve traffic performance in many scenarios, especially unsignalized intersections. However, as for unsignalized intersections where both CAV and Human-driven Vehicle (HDV) exist, to reduce the impact of HDV uncertain behaviors, existing related research tend to simplify the scenario or HDV behavior characteristics at unsignalized intersections, or ensuring passing safety without collaborative decision-making on traffic efficiency optimizations. To address these problems, this paper presents a method of reservation-prioritization-based mixed-traffic cooperative control at unsignalized intersections. Firstly, reservation rights of CAVs are prioritized by solving minimization problems, to optimize the reservation order of CAV on the behalf of traffic efficiency. Secondly, a reservation and speed planning mechanism considering HDV behaviors is designed, which develops and re-decides CAV's reservation result based on HDV free-driving behavior, and plans speed for CAVs based on their reservation results by solving constrained nonlinear programming problems. The proposed method is evaluated under different traffic volumes and CAV penetration rates on SUMO platform. Results show that the proposed reservation-prioritization-based method gains higher intersection throughput and averaged velocity under all scenarios, including a maximum throughput improvement rate of 17.77% and a maximum averaged velocity improvement rate of 66.37% compared with the comparative methods.