Research On Path Tracking Control Method For Automatic Parking System Under Uncertain Interference Conditions

With the increasing number of car ownership and parking space limitations,parking difficulties are gradually becoming a major problem for vehicle users.Automatic parking can not only alleviate the parking difficulty problem,but also reduce parking collisions.The real-time nature and tracking accuracy of path tracking determine the success of automatic parking and chemical material dispatching vehicle scheduling.The existing automatic parking system and chemical park dispatching vehicles,which ignore the outside interference and vehicle steering modeling uncertainty in the tracking control process,lead to the problem of path tracking accuracy degradation,and the path tracking control method considering parameter uncertainty and outside interference is proposed as follows:Firstly,the vehicle kinematic model based on Ackermann steering is established,and Simulink is used to simulate the vehicle motion state under low-speed conditions.The kinematic model is validated by comparing with the vehicle motion data collected from the real vehicle test at low speed and analyzing the process of motion path curvature change.Secondly,according to the performance requirements of automatic parking path planning,according to the characteristics of parking spaces,vehicle kinematics constraints,collision analysis,the geometric method with rapid solution speed is used to carry out the preliminary path planning of parallel and vertical parking,and the corresponding feasi ble parking starting area range is calculated.On this basis,the preliminary paths are smoothed and optimized using higher-order polynomials to obtain the optimized parking paths with continuous curvature and no collision.The feasibility and smoothness o f the path planning algorithm are demonstrated by simulation.Then,a path tracking control method combining expansive state and sliding mode control is proposed for the problem of path tracking accuracy degradation caused by ignoring external disturbances and vehicle steering modeling uncertainties and delays in the actual parking process.The vehicle model considering the uncertainty disturbance is established by introducing compound disturbance,and the corresponding extended state observer(ESO)is established according to the vehicle kinematics to estimate the external disturbance and steering uncertainty of the system.The observed values are used as the compensation quantity in the sliding mode controller,and the path tracking controller applying the disturbance observer is designed for improving the accuracy of path tracking.The effectiveness of the tracking effect of the system was verified by Matlab/Simulink software simulation,compared with the conventional sliding mode control without an observer.Finally,a real-vehicle verification platform is constructed and different parking scenarios are set to test the proposed sliding mode path tracking controller based on the dilated state observer.The experimental results show that the proposed method can achieve good tracking accuracy and control effect under uncertain perturbation parking scenarios.