Responses of Preview-based Vehicle Suspension System on Discrete Impact Roads

The preview-based suspension system improves the performance of the vehicle chassis by adjusting the dynamic characteristics of the suspension in real-time according to the roughness information of the road ahead. In order to reduce the impact load on the vehicle when passing through a discrete impact road, this study proposes a preview suspension control system and control method by combining vehicle dynamics and computer vision. The proposed preview suspension control system acquires the characteristic information of the discrete impact road in real-time. As the main controller, a semi-active suspension controller integrates the real-time road information of the vehicle and performs hybrid state control on the suspension, is proposed. As the front-end previewer, a discrete impact road previewer based on a convolutional neural network is proposed with a YOLO v5s-based discrete impact road target detection algorithm, which can detect and measure the typical characteristic road ahead through computer vision. A simulation platform based on MATLAB/Simulink is established and the ride comfort simulation and analysis of preview control suspension on discrete impact road are conducted. The results show that the vision-based road preview method can accurately detect the characteristic information of discrete impact road surfaces, and the proposed preview suspension control system and method can effectively improve ride comfort and suspension hysteresis under typical road surfaces.