Research On Intelligent Driving Trajectory Tracking Control Algorithm

In recent years,as a cross-cutting research area between traditional technical fields and artificial intelligence fields,intelligent driving cars have received the enthusiastic attention of many companies and scholars worldwide.With the increasing maturity of 5G technology,the penetration of intelligent driving in daily life will also gradually increase.Trajectory tracking control,as the last step to realize intelligent driving,played a decisive role in the smooth implementation of unmanned driving.In the research field of trajectory tracking problem,the tracking algorithm based on geometric principle makes use of the relative position relation between the vehicle and the road in the process of driving,controling the vehicle's front wheel angle according to the reference path information and the current running state,making vehicle move to the next reference point,to realize trajectory tracking control.Because the principle of this method is clear and easy to understand,it is applicable to most vehicle models,so it is widely used in the field of track tracking control.This paper takes the model as the research background of trajectory tracking control algorithm,and designs the relevant control algorithm with the aim of tracking reference trajectory with high precision and high efficiency.The trajectory tracking control system based on geometric principle takes the vehicle's current running speed and preview distance as the control input and the lateral displacement error between the vehicle and the reference path as the system output.The lateral displacement error is an important criterion to measure the tracking effect.The preview distance between the current point and the next reference point is one of the important parameters of the algorithm.The trajectory tracking control algorithm based on geometry takes the current speed of the vehicle and the forward looking distance as control inputs,and uses the lateral displacement error of the vehicle and the reference path as the output.The magnitude of the lateral displacement error is an important criterion for measuring the tracking effect.The forward-looking distance is one of the important parameters.Whether the forward-looking distance is set properly or not will have a direct impact on the tracking effect.This article obtains the vehicle state such as vehicle speed and yaw rate change rate through the vehicle's on-board sensor.Obtaining the error between the vehicle and the desired trajectory at the same time to determine the current trajectory tracking effect,and further design a fuzzy control to choose a reasonable forward looking distance value under different trajectories according to the current state.The tracking control algorithm based on geometric principle does not consider the dynamic characteristics of the vehicle when building the model.In order to solve this problem,the relationship between lateral displacement error and tracking speed is deduced based on vehicle dynamics constraints by establishing vehicle dynamics models such as tire model.combining them with trajectory tracking algorithm.The driving speed obtained from the derivation is guaranteed to meet the requirements of tracking accuracy and the dynamic characteristics of the vehicle.Finally,in MATLAB/Simulink environment,a multi-track compound reference trajectory was first established,and a trajectory tracking model and a vehicle model were established to verify that the control algorithm designed in this paper can achieve high-precision and high-efficiency tracking under complex road conditions.At the same time,a 15-degree-of-freedom vehicle model is established in AMEsim,and the tracking effect of a complex vehicle model is verified under the same conditions.Simulation results prove that the trajectory tracking control algorithm designed in this paper is also applicable to high-precision vehicle models.The development of intelligent driving trajectory tracking technology has certain engineering significance.