Research On Intelligent Vehicle Lane Change Decision And Tracking Control

With the development of science and technology and the improvement of people’s living standards,people’s demand for vehicle autonomy and intelligence is gradually increasing,and smart vehicle technology is also developing steadily.In the face of a complex traffic environment,smart vehicles will make autonomous decisions on their driving behavior and realize autonomous driving functions.Lane-changing behavior is one of the most common driving behaviors.Compared with the driver’s operation,the lane-changing behavior of intelligent vehicles has obvious advantages in terms of safety,comfort,and efficiency.The research work of this paper belongs to the research category of vehicle assisted driving technology.It focuses on the decision-making,planning,and execution of the intelligent vehicle’s constant-speed lane-changing behavior in a dynamic environment.The following are the main contents of this article:(1)Research on decision-making of intelligent vehicle lane changing.First,analyze the vehicle’s lane-changing motivation and main influencing factors.By extracting lane-changing attributes from the real driving data set NGSIM,the vehicle lane-changing behavior decision table is established.According to the high-dimensional nonlinear characteristics of the sample data,a support vector machine decision model is designed.Then the gray wolf optimization algorithm is used to optimize the parameters and kernel function factors in the model.The results show that,compared with the previous model,the improved model has improved accuracy and can implement lane-changing decisions based on the vehicle’s status and environmental information.(2)Research on lane-changing trajectory of intelligent vehicles.Through the comparative analysis of several commonly used lane-changing reference trajectory models,the fifth-degree polynomial is used as the lane-changing reference trajectory.The longitudinal displacement corresponding to different lane changing times is taken as the key point of lane changing,and a series of lane changing reference trajectory clusters are obtained.The comfort is evaluated subjectively and objectively,and the evaluation function is established and transformed into a multi-constraint optimization problem solution,which is optimized by bats.The algorithm solves it to obtain the optimal lane change time,and determines the optimal lane change reference trajectory.(3)Research on Intelligent Vehicle Lane Change Tracking Control.Firstly,a model predictive controller based on a two-degree-of-freedom dynamic model based on the assumption of a small deflection angle is designed.Considering the corresponding constraints,the objective function is established and transformed into a quadratic programming problem to solve the problem.As the change of the front wheel deflection angle at the next moment;afterwards,by analyzing the influence of the prediction time domain and control time domain in the model prediction controller on the tracking effect,an adaptive model prediction controller based on the yaw angle is established,and it is established with Stanley The controllers are compared,and the results show that the improved controller has a smaller tracking error,indicating that the controller can effectively realize the task of lane changing for smart vehicles.(4)The merging model of the road merging area is established,and the merging vehicles in the NGSIM data set are selected to conduct experiments from the lane-changing decision model,trajectory planning model,and tracking control model.The results show that the algorithm in this paper can realize the autonomous lane changing behavior of intelligent vehicles,and the method proposed in this paper meet the requirements of efficiency,comfort and safety in the lane changing process.