Research On Autonomous Lane Change Control Algorithm For Unmanned Vehicles Based On Internet Of Vehicle

The development of driverless vehicles can effectively reduce traffic accidents,relieve traffic pressure and improve traffic efficiency,which is the development trend of future automobiles.Autonomous lane change is a research that must be involved in realizing unmanned cars,and it is one of the key contents of unmanned vehicle research.Lane change decision-making,trajectory planning and trajectory tracking are the three major technologies for autonomous lane change of unmanned vehicles,which will be thoroughly studied in this paper.Firstly,the architecture of the Vehicle-to-Everything is introduced,and the main functions and pros and cons of perception technologies such as cameras,lidar and sensing of the Internet of Vehicles are briefly analyzed.In this paper,the autonomous lane change behavior of unmanned driving in the highway environment is studied,and the lane change behavior is divided into two forms: forced lane change and free lane change,and the minimum lane change safety distance model based on the braking distance of the car is introduced.Secondly,the general high-speed working conditions are simplified,and the free lane change and forced lane change scenarios are established,considering the influence of traffic vehicles in the current lane and the target lane on the lane change safety,and the lane change safety distance model of the current lane and the target lane is established respectively.The speed retention factor and distance retention factor affecting the vehicle lane change decision are analyzed,the information sharing platform of the Vehicle-to-Everything is introduced,and the real-time decision model of free lane change and forced lane change is established based on fuzzy logic theory,and the decision model is simulated and verified.Then,the research on trajectory planning under different working conditions of highspeed scenarios is analyzed.On simple barrier-free roads,Bezier curves are used to fit the lane change trajectory of driverless vehicles.When there are traffic vehicles or obstacles on the driving road,the five-degree polynomial curve is used to describe the lane change trajectory,the trajectory optimization function with improving safety and traffic efficiency as the optimization index is established,the vehicle dynamics constraint is considered,the particle swarm algorithm is used to solve the optimal solution of the trajectory optimization function,and the determination of the evaluation index is analyzed.Introduce the Vehicleto-Everything to provide real-time information to respond to emergency situations,and timely feedback to the decision-making unit for re-trajectory planning.Finally,the vehicle dynamics model is established and simplified,combined with the small angle assumption of the front wheel rotation angle,the prediction model of the front wheel rotation angle as the control quantity is derived,and the nonlinear model prediction controller is discretized.Considering the complexity of longitudinal speed control,a longitudinal velocity control model based on PID is established.In order to ensure the realtime and stability of vehicle trajectory tracking,the nonlinear vehicle dynamics model is linearly discretized,considering various constraints such as road surface conditions and vehicle dynamics,the cost function is designed under the premise of ensuring the accuracy and smoothness of trajectory tracking,and the model prediction controller is established to control the driving direction of unmanned vehicles,and the trajectory is tracked and simulated on the joint simulation platform of Carsim and Simulink to verify the accuracy of the model predictive control algorithm.