Research On Active Collision Avoidance Path Planning And Tracking Control Of Autonomous Vehicle

In recent years,the emergence and development of automatic driving technology has greatly reduced the traffic accidents caused by improper operation of motor vehicle drivers.As an important research direction of autonomous vehicle,active collision avoidance technology has received extensive attention from domestic and foreign automobile manufacturers and scholars,which is of great significance for reducing traffic accidents,improving driving efficiency and ensuring driving safety.At present,the development of vertical collision avoidance systems is relatively mature and has completed commercial implementation.The horizontal collision avoidance system still needs further development.This article focuses on the path planning and tracking control technology of horizontal collision avoidance systems,solving the active collision avoidance problem in structured and unstructured road scenarios.The main research content is as follows:Taking the front wheel steering vehicle as the research object,a path tracking controller based on the model predictive control algorithm is proposed.Taking the vehicle dynamics model as the prediction model,tracking error and control increment error are added to the objective function to ensure that the autonomous vehicle has a high tracking accuracy while consuming the least energy.Various constraints such as control amount,control increment,lateral acceleration and sideslip angle of the center of mass are added,It ensures that the autonomous vehicle meets the mechanical structure constraints of the vehicle when tracking the target path,and has good stability and comfort.For structured roads,a multi-stage lane change collision avoidance path based on quintic polynomial is designed.The coefficient of quintic polynomial is determined by the position of the starting point of lane change,and the lateral velocity and lateral acceleration constraints.The longitudinal lane change position is determined by the lateral safety constraints.In order to ensure that autonomous vehicle have sufficient stability and comfort in the process of lane change collision avoidance,lateral acceleration constraints are added.For unstructured roads,a two-level collision avoidance control system based on model predictive control is designed.The planning layer with obstacle avoidance function plans collision avoidance paths,and the path tracking layer tracks the planned paths.To solve the problem of large system computation,the planning layer uses a point mass model,and the tracking layer uses a three degree of freedom dynamic model,which improves computational efficiency while ensuring control performance.A joint simulation platform of MATLAB/Simulink and Car Sim was established to verify the effectiveness of the designed active collision avoidance path planning and tracking control algorithm under different vehicle speeds and road adhesion coefficients.Finally,the effectiveness of the designed dual layer collision avoidance control system on straight roads was verified using the modified vehicle of BAIC New Energy’s 2016 EV160 as a test platform.The test results show that the autonomous vehicle can effectively perform active collision avoidance at different speeds.During the active collision avoidance process,the yaw angle changes smoothly.The minimum roll angle of the vehicle is-0.18 deg,and the maximum is 1 deg.It has good handling stability and comfort.