Multilane Signal-Free Intersection Cooperation Scheme for Connected and Automated Vehicles with Local Dynamic Resequencing Strategy

This paper studies a new vehicle cooperation scheme for connected and automated vehicles (CAVs) at a multilane signal-free intersection. Firstly, the collision set for each vehicle is constructed, which contains all conflict vehicles that need to be considered by the host vehicle. Then, the cooperation problem of CAVs is formulated into a multiobjective optimization problem to improve traffic mobility, ensure driving comfort and reduce energy consumption according to a global crossing sequence. Thirdly, we propose the local dynamic resequencing (LDR) strategy and vehicle travel time searching algorithm to determine the most efficient global crossing sequence. Specifically, using the LDR strategy, we can obtain the list of new arriving vehicles in the control zone during a fixed time and calculate all the crossing sequences of these vehicles. To evaluate the crossing sequences, we give the vehicle travel time searching algorithm, by which the most effective crossing sequence with the lowest total travel time will be selected. Finally, the simulation results show that the proposed scheme with the LDR strategy can decrease the average travel time by 42.99%–81.66% and reduce fuel consumption by 5.32%–60.52%, in contrast to the no-control (NC) strategy when the range of vehicle arrival rate increases from 360 to 900 vehicles/h/lane. In comparison with the dynamic resequencing (DR) strategy, the LDR strategy can guarantee the fairness of vehicle traffic. In addition, the proposed scheme with the LDR strategy has a good computational performance compared with the first-in-first-out (FIFO) and model predictive control (MPC) strategies and can ensure the safety of vehicle traffic effectively.

Intersections are the main bottlenecks for urban traffic and have caused huge losses to society. The emergence of connected and automated vehicles provides new opportunities to solve the problem. Through vehicle-to-vehicle and vehicle-to-infrastructure communication, the connected and automated vehicles can obtain other vehicles’ state information (speed, position, and so on) and traffic environment information to optimize their trajectories and improve the efficiency of the overall intersection system. This paper proposes a new scheme to regulate connected and automated vehicles to cross a signal-free intersection safely and efficiently. We transform the vehicle collaboration problem into a multiobjective optimization problem to improve traffic mobility, ensure driving comfort, and reduce energy consumption. The simulation results showed that the proposed scheme can reduce the stopping and idling behaviors of vehicles and thus improve the efficiency of vehicle traffic as well as reduce energy consumption. In addition, we propose an algorithm to guarantee the fairness of vehicle traffic, which has rarely been mentioned in previous studies. Although the intersection scenario cannot meet the hypothetical requirements due to technical limitations at present, with the development of technology, the scheme proposed in this paper will have great application prospects in the future.