Study on the Evolution Mechanism of Lane Change Decision in Urban Expressway Diversion Area

Abstract

Urban expressway is the artery of modern urban traffic network, and the vehicle operation at off-ramp and diverging area affects the operation efficiency of the whole traffic system. The lane-changing game behavior between off ramp vehicles and going straight vehicles is very important in the whole driving behavior, and the lanechanging behavior between vehicles is easy to cause traffic accidents and stopping phenomena. To improve the driving efficiency in the diversion area and reduce the accident risk, by analyzing the process of the lane-changing behavior at off-ramp, this paper establishes a two-vehicle game model, solves the replicator dynamic equation according to the Dynamic Evolutionary Game Theory, uses MATLAB to calculate the evolution process and evolution speed based on different payoff factors, explores the influence of safety and speed on the stability of turn-out, and judges the evolutionary equilibrium point according to the determinant and trail of the Jacobi matrix. We build a realistic turn-out scenario and simulate it using the micro-traffic simulation software SUMO and it is found that: (1) The speed of different evolutionary equilibrium points based on speed payoff increased by 8.3% and 4.4% respectively compared with the speed of initial point. (2) The number of conflicts at the evolutionary equilibrium point based on the security payoff reduced to 22% of the initial point. (3) Compared with the initial point, the speed of the evolutionary stable point based on comprehensive payoff increased by 10.3%, and the number of conflicts reduced to 11% of the initial point. The simulation results show that the strategy of stable point of the evolutionary game model can effectively reduce the accident rate and improve the road operation efficiency.