Self-Tuning Optimal Control of an Active Suspension

The objective of this paper is to develop a self-tuning optimal control of an active suspension. An active suspension composed of an identifier and a controller is proposed in this paper. Although control strategies on active (or semi-active) suspensions have been investigated during the past few decades, some problems are not well understood yet. One of them arising from the ride control of an active suspension is that when the weight and the moments of inertia of the sprung mass are varied, the feedback gains of the controller should vary with the variation of parameters accordingly. Therefore, the identifier is proposed before the controller is designed. In the real situations, the parameter variation may occur when loadings on vehicles vary - either from passengers or payloads, especially, in the case of loading on a truck. An identification structure using parallel model reference adaptive system (MRAS) is proposed to identify the true parameters. Figures and numerical data will be presented as a comparison between using a self-tuning optimal control (with parameter identification) and an optimal control (from nominal parameters). Improvements on vehicle responses are expected.