LPV/H∞ suspension robust control adaption of the dynamical lateral load transfers based on a differential algebraic estimation approach

Abstract

This paper is concerned with a new global chassis strategy combining the LPV/H∞ control framework and the differential algebraic estimation approach. The main objective is to enhance the vehicle performances by adapting its control to the dynamical lateral load transfers using a very efficient algebraic dynamical behaviour estimation strategy. Indeed, the lateral load transfers influence considerably the vehicle dynamical behaviour, stability and safety especially in dangerous driving situations. It is important to emphasize that the dynamical load transfers are different from the static ones generated mainly by the bank of the road. The computation of these dynamics must be based on the effective lateral acceleration and roll behaviour of the car. Such effective data cannot be given directly by the hardware sensors (which give correlated measures).The information on the real dynamical lateral load transfers is very important to ensure a good adaptation of the vehicle control and performances to the considered driving situation. A very interesting differential algebraic estimation approach allows to provide the effective needed measures for the control strategy using only sensors available on most of commercial cars. It is based on the differential flatness property of nonlinear systems in an algebraic context.Then, thanks to this estimation approach, the dynamical lateral load transfers can be calculated and used to adapt the vertical performances of the vehicle using the LPV/H∞ for suspension systems control. Simulations performed on non linear vehicle models with data collected on a real car are used to validate the proposed estimation and control approaches. Results show the efficiency of this vehicle control strategy.