Research On Automatic Parking Control Scheme Based On Model-free Adaptive Contro

The emergence of automatic parking system not only reduces the difficulty of parking,but also reduces the occurrence of parking accidents to a certain extent.Most of the tracking schemes use the scheme of controlling the heading angle equal to the expected trajectory slope to realize path tracking,but in the actual tracking process,the actual parking path may be parallel to the expected path rather than coincide,resulting in collision with the vehicles in the front and rear parking spaces in the parking process.Aiming at this problem,this paper introduces a preview-deviation-yaw tracking scheme,and designs the automatic parking control scheme to suitable for two-wheel drive(2WD)and four-wheel drive(4WD)vehicles according to the automatic parking steps.For 2WD vehicles,this paper designs a preview-deviation-yaw tracking control scheme based on model-free adaptive control(MFAC).The coordinate information and body angle of the vehicle at each moment are used in calculation the preview-deviation-yaw,the parallel rather than coincident situation between the actual parking track and the expected track can be avoided in the process of tracking,and can improve the accuracy of path tracking.For4 WD vehicles,a preview-deviation-yaw scheme based on proportional integral full form dynamic linearization based model-free adaptive control(PIFFDL-MFAC)is proposed in this paper.At the same time,the corresponding control algorithm,parameter estimation algorithm and parameter reset algorithm are given,and the stability of PIFFDL-MFAC control algorithm is analyzed and proved.In order to verify the effectiveness of the design scheme in this paper,the simulation experiments of 2WD automatic parking control scheme and 4WD automatic parking control scheme are carried out in MATLAB environment.The 2WD automatic parking system is a single input single output(SISO)system with the front wheel angle as the input and the preview-deviation-yaw as the output.PID tracking scheme and Fuzzy tracking scheme are used to compared with MFAC tracking scheme.4WD automatic parking system is a multi input single output(MISO)system with the speed and angular velocity of the vehicle geometric center as the control inputs and the preview-deviation-yaw as the control output.The PIFFDL-MFAC control scheme is compared with the conventional FFDL-MFAC control scheme and PID control scheme.The 2WD simulation results show that three tracking control scheme can complete the automatic parking process for different vehicles.And can be seen that the tracking effect of MFAC tracking scheme is better than PID and Fuzzy tracking scheme by further comparison and analysis of tracking error and heading angle error generated in the parking tracking process.The simulation results of 4WD show that three tracking control scheme can complete the automatic parking process for different vehicles.Due to the output error proportional control term is added,the PIFFDL-MFAC control scheme can work on the change of system output and reference signal,and has faster corresponding speed and higher control sensitivity than conventional FFDL-MFAC control scheme.Moreover,it has smaller tracking error than PID control algorithm.