Research On Intelligent Vehicle Automatic Lane Change In Expressway Environment

The development of intelligent vehicles has great potential in reducing traffic accidents and relieving traffic pressure,which has become a research hotspot in the world.Automatic lane change is a research which must be involved in the realization of intelligent vehicle,and it is one of the key contents of intelligent vehicle research.Research on automatic lane change of intelligent vehicles is of great significance to improve road traffic safety and efficiency.In this paper,the automatic lane change of intelligent vehicles is studied under the highway structured road.The lane-changing behavior is divided into two forms: forced lane-changing and free lane-changing.Different lane-changing scenes and lane-changing processes are studied and analyzed.The movement state of main vehicle and traffic vehicle is analyzed,and the safe distance model of current lane and target lane is established.In the process of establishing the decision-making model of lane-changing,by analyzing the factors affecting the decision-making of free lane-changing and forced lane-changing,the speed difference value coefficient and the distance expectation are taken as the input of the decision-making model of free lane-changing,the speed of self-driving and the distance from the latest time of lane-changing are taken as the input of the decision-making model of forced lane-changing,and the willingness of lane-changing is taken as the output.Based on the theory of fuzzy logic,the decision models of free lane-changing and forced lane-changing are established respectively,as the initial decision-making of lane-changing.Simulation experiments based on CarSim are carried out to verify the rationality of the two decision-making models.In the trajectory planning method,the polynomial-based trajectory planning method is selected,and the rolling window optimization algorithm is introduced.The trajectory is updated repeatedly on-line with the time as a cycle to meet the real-time requirement of lane change.A trajectory optimization function with vehicle lateral acceleration and lane-changing time as optimization indexes is introduced to select lane-changing trajectories in different planning periods,and the influence of different weight ratios of optimization indexes on lane-changing is analyzed.In the dynamic constraints of lane-changing trajectory,the driving conditions are divided into the limit driving conditions and general driving conditions.Underthe limit driving conditions,the minimum turning radius of vehicles in ideal state is discussed,and the minimum turning radius of vehicles in ideal state is verified by simulation experiments,and the actual minimum turning radius of vehicles in different road conditions is obtained.Under general driving conditions,the dynamic model of vehicle tire is analyzed,and the minimum turning radius under general driving conditions is determined.By restricting the minimum turning radius of vehicles,curvature of lane change trajectory is limited to prevent side slip and other dangerous conditions when vehicles change lanes at high speed.Finally,the trajectory planning under different working conditions is simulated and validated by using MATLAB and CarSim software.The simulation results show that the real-time dynamic trajectory planning algorithm based on rolling window optimization meets the requirements of real-time and security of lane-changing.Dynamics simulation results show that the planned trajectory meets the requirements of vehicle dynamics and can realize dynamic and safe Lane change.