Intelligent Driving Autonomous Overtaking Dynamic Avoidance Planning And Tracking Control

The behavior of overtaking is a common traffic behavior,which is strictly regulated in traffic laws.In the field of intelligent driving,the safety of autonomous overtaking is also one of the urgent problems to be solved.In this paper,the overtaking behavior of intelligent vehicle is mainly divided into three steps:lane change,overtaking the target vehicle and changing back to the original lane,and the decision planning,longitudinal and lateral control of intelligent vehicle are studied respectively.In the aspect of lane change decision planning,the position state that the intelligent vehicle needs to meet with the obstacle vehicle in order to avoid collision at the beginning of lane change was analyzed.A safe and smooth quintic polynomial lane change trajectory was screened by predicting the influence of occupancy grid information and lateral acceleration on driving stability.In this paper,a dynamic quintic polynomial lane change trajectory planning algorithm was proposed because of the real-time change of obstacle vehicle states in the environment.In the aspect of longitudinal control,the collision types that the intelligent vehicle may have with the obstacle vehicle in the environment when overtaking the target vehicle were analyzed.According to the constraints of obstacle vehicles ask of the intelligent vehicle on the position and speed.This paper proposed a longitudinal controller design method based on a division of safety threshold line.And a longitudinal controller was designed by double-layer fuzzy inference system controller and fuzzy PID controller.In the aspect of lateral control,this paper introduced the basic principle of model predictive control.And the two-degree of freedom intelligent vehicle dynamics model and the intelligent vehicle error state equation was established.To utilize the objective function combined with the constraint conditions of road designed a lateral controller based on model prediction control in order to realize the track of overtaking track.Finally,the autonomous overtaking algorithm of this paper was verified,through Carsim and Simulink platform,when the intelligent vehicle has different initial speeds and the obstacle vehicle has abnormal driving behavior.Simulation results show that the algorithm has the traits of real-time and effective.