Research On Traffic Simulation Environment Construction And Efficiency Evaluation Of Cooperative Vehicle Infrastructure System

With the augment of Chinese vehicle ownership and the enhancement of motorization,the traffic problems in big cities are increasingly becoming obstacles to the development of cities and the improvement of people's living standards.Real-time adjustment of vehicle operating conditions based on cooperative vehicle-infrastructure system as an effective means of solving urban traffic problems have received much more attention.Cooperative vehicle-infrastructure technology mainly contains access to information about vehicles and roads based on the wireless communication technologies and carries out information exchange through V2X(vehicle to everything)technology so as to share resources,further optimize system resources effectively,improve road traffic safety,ease traffic congestion,and improve traffic efficiency.In the thesis,we will build the simulation environment of the cooperative vehicle-infrastructure system,establish the traffic efficiency evaluation index system,and evaluate the traffic efficiency of the cooperative vehicle-infrastructure system based on the multi-index decision-making method.The research work accomplished in the thesis mainly contains:(1)To begin with,the current research on the traffic efficiency and the key indicators of cooperative vehicle-infrastructure system are introduced on the basis of the development of cooperative vehicle-infrastructure system.The thesis analyzes the operation mechanism of positioning error,communication delay and penetration rate in the simulation of the key indicators of the cooperative vehicle-infrastructure system.Based on the traffic efficiency metrics,the thesis further selects the traffic efficiency evaluation indicators from the intersection,road sections and road network.(2)The thesis constructs a complete index system of traffic efficiency evaluation,then using the multi-index decision method to construct AHP-BP neural network in consideration of the huge number of traffic evaluation indexes method.The thesis firstly determines the weights of traffic parameters based on AHP and gets the results of the evaluation.Then the results are further trained through the BP neural network.Finally,the training results can be directly applied to the analysis of traffic examples of different roadway coordination systems and can be directly obtained by using the model to save computation.(3)At the same time,the thesis uses TOPSIS method based on double-point weighting to objectively determine the weight of the evaluation index on basis of the objective data,then combining the similarity measure of preference similarity and negative preference similarity.The superiority and inferiority of traffic efficiency in different cases are compared by the obtained degree of closeness,ensuring the objectivity of the result.The thesis uses Q-paramics traffic simulation software and VS2013 to construct the simulation environment of cooperative vehicle-infrastructure system.Both traffic simulation tests and evaluations are carried out in three scenarios based on CVIS and common traffic system respectively.The first is to simulate the traffic simulation under different road network scale comparing the corresponding traffic efficiency under different traffic loads;The second is to take consideration of the index of CVIS:positioning error,communication delay,and penetration rate on traffic efficiency;The third is to deeply evaluate the traffic efficiency of CVIS and common traffic system under the influence of the special environment(accident led to traffic congestion,intersection congestion,signal failure,etc.)The thesis evaluates the traffic efficiency of different instances in different scenarios.The results show that the CVIS within the range of the key indexes can effectively improve the traffic efficiency.