Research Of Driver's Route Choice Behavior Based On Evolutionary Game

With the rapid development of urban economy,road congestion,traffic accidents and other traffic problems occur frequently.In urban traffic system,because of the macroscopic characteristics of road network and the complexity of road information,drivers are unable to receive real-time and dynamic road information in an all-round way.Therefore,when the traffic management department provides guidance information,it will inevitably lead to a game between drivers and traffic managers.At the same time,there will also be a game conflict among drivers of different groups.In order to maintain good traffic order and alleviate the traffic pressure of urban roads to a certain extent,this paper uses evolutionary game model to conduct in-depth research on driver's route choice behavior,which mainly includes the following aspects:(1)Based on the traffic guidance information and the different psychology of drivers in choosing the route,the evolutionary game model between traffic managers and drivers,and the evolutionary game model of the route choice behavior among different driver groups are established respectively.(2)On the basis of the two evolutionary game models,according to the idea of game theory,the profit matrix of the game among drivers and traffic managers and the profit matrix of the game among drivers of different groups are obtained.The replication dynamic equation is solved for the two evolutionary game models respectively,and the replication dynamic system is obtained.In the analysis of the stable state of system evolution,the Lyapunov's first method in the control system theory is applied.Combining with the operation of Jacobian matrix and the properties of determinant,the stability conditions are reasonably assigned to each parameter in the established interval,and the final evolutionary steady state is generated.Based on the traditional replication dynamic equation,this paper considers the influence of experiential learning rate on the driver's route choice behavior,and then extends the original replication dynamic equation.Taking 2x2 game as an example,the elastic influence factor is introduced for quantitative analysis.The improved replication dynamic equation is applied to the game model of drivers in two groups,and the final evolutionary stable solution and evolutionary stable strategy are obtained.(3)Under the condition of local stability,this paper simulates the equilibrium points in two evolutionary game models by using Matlab,and obtains the corresponding replicated dynamic phase diagrams,which reflects the evolutionary trend of each stable point intuitively.The evolutionary stable state before and after the improved replication dynamic equation is compared,and the stimulating and blocking effects of the elastic influence factor in the same driver group are verified.That is to say,if the experiential learning rate of the driver choosing one path is greater than that of the other,then choosing this path will hinder the choice of another path.At this time,the stable state of the system will be better when choosing this path,and it will be imitated in the driver's game group.