Improving Vehicle Handling and Stability under Uncertainties using Probabilistic Approach

Abstract

In this paper, we consider improving vehicle handling and stability of road vehicles under model uncertainties using a probabilistic robustness approach. Model based approaches have been extensively explored to tackle the handling and stability problem. However it is known that the performance under such techniques can degrade if uncertainties in model are not taken into account. Many severe accidents result from loss of stability directly or indirectly, which may be attributed to emergency steering, tire pressure loss and different road surface adhesion in the vehicle model. One approach to cater for uncertainties is optimal guaranteed cost controller (OGCC), X. Yang, Z. Wang, and W. Peng [2008], which follows the worst case methodology of robust control design. In this paper, we aim to enhance the performance of the system by formulating the uncertainties as a stochastic phenomenon endowed with a probability measure. Inspired by recent major result in probabilistic convex optimization, we guarantee a priori probabilistic robustness while allowing for certain violation of the underlying constraints of the optimization problem. Numerical simulation results show a marked improvement in results in comparison to the optimal cost (OC) and OGCC approaches.