Active Suspension Control Based on Estimated Road Class for Off-Road Vehicle

Abstract

Road input can be provided for a vehicle in advance by using an optical sensor to preview the front terrain and suspension parameters can be adjusted before a corresponding moment to keep the body as smooth as possible and thus improve ride comfort and handling stability. However, few studies have described this phenomenon in detail. In this study, a LiDAR coupled with global positioning and inertial navigation systems was used to obtain the digital terrain in front of a vehicle in the form of a 3D point cloud, which was processed by a statistical filter in the Point Cloud Library for the acquisition of accurate data. Next, the inverse distance weighting interpolation method and fractal interpolation were adopted to extract the road height profile from the 3D point cloud and improve its accuracy. The roughness grade of the road height profile was utilised as the input of active suspension. Then, the active suspension, which was based on an LQG controller, used the analytic hierarchy process method to select proper weight coefficients of performance indicators according to the previously calculated road grade. Finally, the road experiment verified that reasonable selection of active suspension’s LQG controller weightings based on estimated road profile and road class through fractal interpolation can improve the ride comfort and handling stability of the vehicle more than passive suspension did.