Application And Research Of Electrical Parking Brake System In Parking Assist

Parking Assist System possesses a great market demand for this system can assist drivers in parking an automobile in confined environments. Meanwhile, the importance of safety in parking is increasingly significant; as the advanced automotive electronic system, Electrical Parking System(EPB) is capable of controlling the vehicle speed effectively, improving the vehicle active safety in parking. For this very reason, EPB is introduced into parking assist; the overall architecture combined with binocular stereovision is designed; the principle of EPB in parking assist is analyzed, and the brake control strategy based on calculus of variations in parking is studied.This paper analyzes the structure and working process of EPB in parking assist; the Computer Vision is used as the parking environment construct method; the image process software is designed for camera calibration; the feature of image in parking environment is analyzed, and the Scale Invariant Feature Transform(SIFT) is selected as the image registration method. By building the cross-compiling environment on PC, the boot-loader, Linux kernel and filesystem are compiled, and then the parking image acquisition system based on ARM and Linux is established. The parking space construction method is studied from the perspective of matrix analysis, and the representative scenes, namely the pedestrian, wall, automobile, parking mark and plant are used to verify the construction method, with the relative errors all being less than 1%.That the sliding friction work of brake and the longitudinal shock extend of vehicle serve as the evaluation for brake process quality is proposed; the dynamic characteristics of transmission system composed by engine, hydraulic torque converter and driving shaft on parking condition is analyzed. Taking the change rate of brake torque as the control variable, the objective functional J is constructed with the weighted sliding friction work of brake and square of control variable based on calculus of variations; combined with the constraint condition while parking, the optimized curve of vehicle speed, used as the reference, is generated; the difference between the reference and the one acquired by wheel speed sensor is fed back to EPB, forming closed loop control; Matlab simulation shows that the vehicle safely ceases with the constraint of certain time and distance. Measuring the related parameters of automobile, the optimized curve of vehicle speed is designed. In real vehicle test, the automobile stops at expected distance, and the feasibility of control strategy is preliminarily verified.