Research on Pattern Recognition of Automobile Anti-Lock Braking System

Anti-lock braking control system is constructed on the basis of traditional braking system by attaching ABS pressure regulating system, which adopts intelligent control technology to regulate the braking force applied to each wheel to prevent wheel locking. In this paper, the pattern recognition of the automobile anti-lock braking system (ABS) is investigated, and simulation experiments of the braking process are carried out for the ABS system. At the same time, in order to explore the ability of ABS system to recognize the wheel speed of the car, this paper carries out simulation experiments for the vehicle with ABS system in the case of no fault and sensor failure, so as to construct the automotive ABS fault pattern recognition model based on BP neural network, and train the network using the Levenberg-Marquard algorithm. The ABS braking control system used in this paper, in the braking process, the automobile wheels did not have the phenomenon of locking, the angular acceleration of the wheels is controlled at -75~75rad/s², and the slip rate can be controlled in the range of 0.2~0.5 where the maximum adhesion coefficient can be obtained. The automobile ABS failure mode recognition model constructed in this paper reaches the predetermined value of 0.0001 error after 400 iterations, and the simulation output and target output during training are very much in line with each other, and the relative error is overwhelmingly less than 3%. It shows that the model can fulfill the target of recognizing ABS failure modes well, and it has a certain inspirational significance for the research of pattern recognition in automobile anti-lock braking systems.