Research On System Control Strategy Of CAV In Mixed Traffic Flow Under Network Connected Environment

Connected Autonomous Vehicle(CAV)technology integrates self-driving vehicle technology and Internet of Vehicles technology,greatly improving the performance of the vehicle,and thus has the ability to optimize driving safety,traffic efficiency,energy saving and emission reduction.With the continuous development of technology,the headway of CAV will continue to decrease and road capacity will continue to improve,and related research has become a current research focus.And CAV can be controlled by the system to reduce the efficiency loss caused,which brings new potential for alleviating traffic congestion.Based on the background of the transition period from HV to full CAV,this thesis analyzes the influence of mixed traffic flow on the capacity of road sections,and builds a variational inequality model of multi-user mixed traffic flow distribution in a networked environment.The algorithm for solving the model is designed based on the idea of diagonalization.Since all the known diagonalization algorithms have a uniform step size,in order to adaptively solve the problem raised,a non-uniform step size determination method based on the automatic adjustment averaging method is proposed.The analysis results of the calculation examples show that the proposed distribution model can more accurately describe the distribution of mixed traffic on the road network.The designed non-uniform step size solution algorithm has the advantage of rapid convergence compared with the existing diagonalization algorithm to solve this problem.At the same time,the analysis of the calculation example shows that the path selection of the CAV in the system control part can bring system gains.From this,the total impedance gain and control strength of the road network are used as the objective function to establish the optimal control ratio model including the overall ratio and the OD pair under the fixed overall ratio.Under the uncertainty of the proportion and the overall proportion,the OD comparative proportion can reasonably determine the number of vehicles controlled by the obedience system to improve the operation of the road network.And based on genetic algorithm and improved particle swarm optimization design algorithm,and combined with examples to verify the effectiveness and accuracy of the model.The results of the calculation example show that the three control methods all reduce the system cost to a certain extent,and OD has the most significant effect on the control method,and the system cost is reduced by about 5%.In terms of the effectiveness of the algorithm,the genetic algorithm,particle swarm algorithm and improved particle swarm algorithm of the large-scale road section are compared to solve the model.The improved particle swarm algorithm converges faster when solving this problem.In order to ensure the homogeneity of the traffic flow and weaken the control strength,it is considered to set up a CAV dedicated lane layout plan.Under the premise of setting up the CAV dedicated road network,analyze the characterization method of the CAV system control intensity,construct the joint optimization model of the CAV system control strategy and the dedicated lane setting plan,and build the multi-dimensional discrete particle swarm algorithm to solve the two-level programming model.The results of the calculation examples show that the proposed strategy can further reduce the system cost and reduce the control intensity.Based on the calculation examples,the proposed method is based on the sensitivity analysis of CAV dedicated lane construction cost limit,CAV market penetration,traffic demand and CAV dedicated lane capacity to determine the key factors affecting the system.Urban traffic network flow distribution problems and control strategies for new travel modes are an important part of studying the balance of urban traffic supply and demand and alleviating future urban traffic congestion.During the transition period from HV to full CAV,the construction and deployment of a joint optimal control ratio strategy for CAV dedicated lanes can make full use of the advantages of CAV and promote the sustainable development of urban traffic.