Research On Path Following Control Of Autonomous Vehicles

Autonomous vehicles are widely used in civil,military,scientific research and so on.The key techniques include path following,path planning and autonomous obstacle avoidance.This paper mainly focuses on the strategy of path following based on autonomous vehicles.Firstly,two kinds of path following control strategies are proposed,namely,the path following strategy based on the desired yaw rate and the path following strategy based on the transverse preview error model.The former is aimed at ensuring the lateral stability of the vehicle,while the latter focuses on reducing the lateral preview error and the orientation error.Then,the direct yaw moment control(DYC)strategy is designed to distribute the four tire longitudinal forces.The simulation results show that the former has better path-following performance and lateral stability.Then,the sideslip angle during the actual vehicle driving process cannot be measured,and the sensors itself has some noise,the Kalman filter is designed to estimate the vehicle sideslip angle and yaw rate when considering the side wind disturbance and the sensor noise.According to the separation theorem,the LQG controller is designed to realize the path following of autonomous vehicles.Aiming at the deficiency of the stability of the LQG controller,the improved LQG/LTR controller is designed to improve the robustness of the system.Finally,the paper analyzes the high speed obstacle avoidance strategy of the vehicle,the polynomial is used to plan the longitudinal uniform polynomial lane changing path.In the case of high speed vehicle,different obstacle avoidance path is designed according to the different longitudinal obstacle avoidance distance.The simulation results show that the proposed path-following strategy can accurately track the planned high-speed obstacle avoidance path.