Study On Intelligent Vehicle Lane Change With Game Theory

In many traffic accidents,the majority of traffic accdients are caused by improperly lane change.Therefore,the ability of lane change in intelligent vehicle has become the hotspot.In this paper,a large number of literatures related to intelligent driving are compared and analyzed.The main process of its work is lane-change decision,lane change trajectory planning and lane-change trajectory tracking control.First of all,lane change has safety requirements.Based on the analysis of relevant lane change scenarios,the shortest safe distance in lane change has prompted.Then the concept of payoff in game theory is used to establish.After that the speed payoff and the comfort payoff were used to evalute vehicle lane change.Intelligent vehicles need the referenceline as the driving benchmark.In this paper,the smooth algorithm,similarity algorithm and compact algorithm are proposed to optimize the referenceline,so that the intelligent vehicle has a good referenceline.At the stage of decision-making,the SL diagram is used to analyze the situation of non-crossing.The ST diagram is used to analysis the situation of crossing.Generate convex space and dynamic programming to meet requirement of lane change.In the aspect of trajectory planning.In order to make up for the shortcomings of quintic polynomial lane change,Bessel curve is used to run rajectory planning.Because of the difference of coordinate system in trajectory planning and control.So the transform between Cartesian and Frenet are needed.In the aspect of trajectory tracking control,the controller is based on the two degrees of freedom of vehicle model.The longitudinal control of the vehicle use double PID which are based on error of speed and position.Discrete LQR control is used to realize lateral control.Finally,through the simulation of Carsim,Prescan and Matlab to verify the algorithm.The result of simulation shows that,in the situation of non-crossing,decision-making in SL can guide vehicle to overtake or slow down.In the situation of crossing,decision-making in ST can make vehicle avoid collion through accelerate or decelerate.In the end,to use payoff of speed and comfort to evaluate lane change.