Autori
Giorgio Battistelli, Daniela Selvi and Alberto Tesi
Abstract
A relevant and challenging problem in control design is that of reducing the effects of undesirable input signals on the output of the process to be controlled. Disturbance attenuation arises in several diversified contexts, e.g., active noise control, eccentricity compensation, active suspension systems, laser systems, disturbance torque compensation in electric machines, disturbance attenuation for adaptive optics applications for ground-based telescopes. In practical situations, a non-negligible uncertainty usually affects the disturbance model, which could, in addition, exhibit time-varying characteristics. Specifically, when such an uncertainty prevents a single robust controller to be able to ensure satisfactory performance, algorithms for real-time control reconfiguration are needed. The interest for adaptive solutions for the problem of disturbance attenuation is witnessed by recent works on the topic. We present our research and results related to this problem.
