Research On Local Trajectory Planning And Tracking Control Algorithm Of Intelligent Vehicle

With the rapid development of artificial intelligence and 5G technology,self-driving technology are recognized and gradually into the people’s field of vision,the development of intelligent driving technology in China is entering a new stage.Intelligent driving is the main research direction of automobile in the future,and it is an important embodiment of the strength of China’s scientific and technological power.More and more enterprises are committed to the research of intelligent driving technology.The research of path planning and tracking control system is the key part of intelligent driving technology.The accuracy of planned path and tracking is an important standard to judge vehicle intelligence.Therefore,this paper will conduct in-depth research on the path planning and tracking control system.By summarizing the advantages and disadvantages of various planning algorithms and tracking control algorithms,the improved A* algorithm is selected for trajectory planning,and the horizontal LQR control algorithm and vertical double PID control algorithm combined with fuzzy control algorithm are selected for tracking control.The main work includes:(1)in-depth study of the current situation of the development of intelligent driving technology at home and abroad,focusing on the car companies,Internet companies,and the gap between the colleges and universities.At the same time,the paper analyzes the intelligent driving in path planning and tracking control technology research status.Through reading materials,analyzes the advantages and disadvantages of each algorithm proposed in this paper,the algorithm.(2)It is difficult to efficiently express the connection between trajectory information and vehicle,and it is difficult to carry out trajectory planning in geodetic coordinate system in three-dimensional space due to the time dimension involved.Frenet coordinate system is adopted to decompose three-dimensional trajectory planning into horizontal and longitudinal two-dimensional problems to solve,and at the same time,it can well describe the connection between vehicle state and trajectory information.(3)An improved A* algorithm was proposed,combined with the driving area model to build A trajectory planning system.The two-step A* algorithm was used to realize the dynamic obstacle avoidance of vehicles.Constraints such as expansion strategy,smoothing cost and collision cost were added to the algorithm to improve the adaptability of local trajectory.(4)Tracking control algorithm of this paper is based on the transverse LQR control algorithm and longitudinal double PID control algorithm to achieve,tracking error model is established in combination with feed forward control structure tracking control algorithm.After building vehicle dynamics model and the tire model,increase the tracking of the adaptive fuzzy control algorithm,after the completion of the tracking control algorithm can accurately track the desired trajectory.(5)The trajectory planning algorithm and tracking control algorithm are verified in the co-simulation and comparison experiments of Prescan,Carsim and Simulink software.The results show that the improved algorithm has advantages in horizontal tracking error,longitudinal tracking error and heading Angle error.Figure [66] Table [10] Reference [78]...