Design of a Predictive Semiactive Suspension System

Abstract

In this paper we present an original design procedure for semiactive suspension systems. Firstly, we consider a target active control law that takes the form of a feedback control law. Secondly, we approximate the target law by controlling the damper coefficient f of the semiactive suspension. In particular, we examine two different kinds of shock absorbers: the first one uses magneto-rheological fluid instead of oil, while the second one is a solenoid valve damper. In both cases the nonlinear characteristics force-velocity of the damper are used to approximate the target law. To improve the efficiency of the proposed system, we take into account the updating frequency of the coefficient f and compute the expected value of f using a predictive procedure. We also address the problem of designing an asymptotic state observer that can be used not only to estimate the current state but also to predict the value that the state will take at the next sampling time.