A Traffic Flow Analyzing Based Vehicle Scheduling Scheme At Road Intersection In VANETs

Nowadays,the rapid development of autonomous vehicles and advancements of VANETs have enabled new methods to solve the problem of increasing traffic congestion and road safety at road intersections due to the rapid increase in urban motor vehicle ownership.Using VANETs instead of traffic lights to dispatch vehicles,improve traffic efficiency at intersections,and conduct accident prediction and collision avoidance is an effective solution.However,in the absence of signal lights,complicated vehicular movement constraints,conflicting vehicle trajectories,and differences in vehicle traffic demand,and many other real-world traffic problems pose serious challenges to the design of vehicle scheduling methods.In order to improve the vehicle travel efficiency and road safety at road intersections,this thesis proposes a new vehicle scheduling method at road intersections based on Connected and Autonomous Vehicles(CAVs).Firstly,the proposed method utilizes VANETs to obtain global traffic status at the intersection area,adds vehicles to the vehicle flow and dispatches vehicles in units of vehicle flow,the movements of vehicle flow are coordinated for reducing the time delay of vehicles at road intersections.Secondly,the proposed method utilizes the path planning of autonomous vehicles to obtain vehicle trajectories,the trajectory conflicts can be detected and vehicle trajectories at conflict points can be modified to avoid vehicle collisions and guarantee the road safety,meanwhile,the capacity of road intersections are improved.Finally,the proposed method gives emergency vehicles higher priority of crossing to allow emergency vehicles to have a faster travel across road intersections.Theoretical analysis proves that the proposed method is deadlock free and starvation free during vehicle scheduling period.A crossroad is set up on the simulation platform as a typical traffic scenario for traffic simulation.To verify the efficiency of the proposed method,a traffic simulation experiment is set up in the simulation platform to implement the proposed vehicle scheduling method.The simulation results show that the proposed method performs better than existing slot based and traffic light based intersection management schemes under different traffic conditions.