New results on robust stability analysis and synthesis for MIMO uncertain systems

Abstract

This paper provides theoretical preliminary results and develops a first methodology, that allows one to efficiently estimate the maximum time constant of a dynamic matrix of an uncertain system with rational multi-affine structure with respect to parameters, and a second one to find a majorant system of a multi-input multi-output (MIMO) system with rational multi-affine structure with respect to parameters. The developed methodologies are used to estimate the evolution of an uncertain linear time-invariant (LTI) system with additional bounded nonlinearities and/or additional bounded input signals. Moreover, the above results are also used to design a robust controller for an uncertain MIMO system with unmeasurable states and subject to a rate-bounded disturbance in order to track a rate-bounded reference signal. The obtained theoretical results are illustrated by three examples. The first two examples deal with the analysis of a LTI system with bounded disturbances and measurement noise, and additional bounded and not bounded nonlinearities, respectively; in the second example a new control law with saturation is also designed. In the last example, a robust controller for an uncertain electro-mechanical system with unmeasurable state is designed to track a rate-bounded reference signal in the presence of a disturbance with bounded derivative.