Research On Trajectory Tracking Control Algorithm Of Autonomous Vehicle In Emergency Situation

In recent years,with the rapid development of AI(artificial intelligence),IOT(Internet of Things),big data and other new-generation electronic information technologies,electrification,connection and intelligence have become the future development direction of automobiles.As the core technology of vehicle intelligence,autonomous driving technology provides a huge guarantee for improving traffic congestion,vehicle safety and ride comfort.This thesis is focuseed on the topic " Research on trajectory tracking control algorithm of autonomous vehicle in emergency situation",and typical issues such as,path planning,trajectory tracking control and vehicle stability control are studied,for realizing obstacle avoidance and vehicle control in emergency condition.The main research contents of the thesis are as follows:First,the research content of the thesis is clarified.Around the topic of the thesis,the research background and significance of the thesis are analyzed.The research status of intelligent vehicle path planning at home and abroad,intelligent vehicle trajectory tracking at home and abroad,and intelligent vehicle chassis stability control at home and abroad are comprehensively described,and the research content and technical route of this thesis are put forward.Secondly,the dynamic path planning algorithm based on Frenet coordinate system is studied.The coordinate transformation relationship between Frenet coordinate system and Cartesian coordinate system is deduced.In Frenet coordinate system,5-degree and 4-degree polynomials are used to plan the horizontal and vertical candidate trajectory sets respectively.Based on the safety distance and edge expansion method,the candidate trajectories are evaluated,and the loss function is used to select the best candidate trajectory.The simulation verification of local avoidance path planning is carried out in the emergency avoidance scene of straight and curved lanes.Thirdly,the trajectory tracking control algorithm and controller design of intelligent vehicle based on improved model predictive control are studied.A three degree of freedom vehicle dynamics model is established as the theoretical model.The tire cornering stiffness is estimated by the ordinary least squares considering genetic factors.The unscented Kalman filter algorithm is used to estimate the yaw rate and sideslip angle of the vehicle.The data fitting method is used to obtain the relationship between the model prediction time domain and vehicle speed and road adhesion coefficient.The weight coefficient of lateral displacement deviation is dynamically adjusted by fuzzy control algorithm.The co-simulation of Car Sim and Simulink is used to verify the effect of trajectory tracking controller.Fourthly,vehicle stability control based on sliding mode variable structure steering differential cooperative control is studied.The nonlinear two degree of freedom vehicle model is established,and the stability region boundary is determined by using the phase plane theory.The active front wheel steering control law is designed based on the inverse dynamic sliding mode variable structure,and the direct yaw moment control law is designed based on the adaptive super spiral sliding mode variable structure.The cooperative control relationship between front wheel active steering and direct yaw moment is designed by smoothing function.Taking the adhesion utilization rate of tire and road surface as the optimization objective,the optimal torque distribution scheme is designed to realize the accurate tracking of the desired centroid sideslip angle and the desired yaw angle.The co-simulation of Car Sim and Simulink is used to verified the effectiveness of the cooperative controller.Fifthly,the path planning,trajectory tracking control and vehicle stability control of intelligent vehicle under emergency avoidance condition are verified by co-simulation.Based on Car Sim and Simulink co-simulation platform,the co-simulation verification is carried out under the conditions of static obstacles and dynamic obstacles of straight lane under high adhesion conditions and static obstacles and dynamic obstacles of curve lane under medium adhesion conditions.The simulation results show that the designed path planning algorithm,trajectory tracking controller and stability controller can ensure vehicle collision avoidance under emergency conditions,and the collision avoidance process is smooth and stable.Finally,the research content of the thesis is summarized and prospected.