Research On Intersection Vehicle Management Under Vehicle-to-vehicle Environment

With the widespread of artificial intelligence,great changes had taken place in the area of intelligent transportation system(ITS).Initial implentation of car networking and autonomous vehicles had laid a good foundation for revolutionary transformation of traffic.Traffic control methods under vehicle-to-vehicle environment had become one of the focuses of people's research.Assuming that the autonomous vehicles had been maturely applied,this thesis studied the vehicle.management methods at intersection.Firstly,a basic method of vehicle management in intersection area was proposed.The information interaction rules of vehicles and intersection control center were defined and vehicle scheduling model was built.Then an objective function that minimizes the time when vehicles passed through the intersection was established,a rolling horizon optimization algorithm was used to solve the model.After that,the method was simulated using AIM platform.Compared with the FCFS strategy,the method of this thesis can improve intersection throughput and reduce the average delay of vehicles.In addition,based on the methods above,a strategy focused on the special vehicles was put forward.The special vehicles were classified into two types,namely emergency vehicles and special-demand vehicles.The former was used an emergency vehicle scheduling strategy to manage the vehicles.The latter was established a collision matrix of vehicles at intersection,then using the interaction rules to reduce vehicles conflicts.After that,an objective function aimed to maximum the number of vehicles involving the scheduling was provided,and the optimal solution was found using Tabu search algorithm.Simulation results showed that the Tabu search strategy increased the throughput compared with FCFS-Emerg and Traffic-Light strategy,the avarage delay was reduced under the same circumstance.Finally,the car-following and lane-changing rules are studied based on cellular automaton.According to the driving characteristics of autonomous vehicles,car following models were improved and a STCA-V lane-changing model was provided.The efficiency and effectiveness was demonstrated based on the cellular automaton using simulation platform,which laid a good foundation for further study.