Control of Linear Time-Invariant Systems with Plant Disturbances and Incomplete Measurements

Abstract

This paper is concerned with the estimation and control of linear systems with unknown disturbances entering the system at specified points. The assumption that systems experience some sort of disturbance input which is not known a priori and which cannot be directly measured on-line is true for most cases. These disturbances may be due to environmental effects or modeling approximations. As examples one can cite load fluctuations in regulators, changes in wind velocity and direction in aircraft; raw material variations in chemical processes, etc. Estimators designed by conventional techniques fail exactly to reconstruct the state when such disturbances are present. In this case the designer is forced to use output feedback or an unknown-biased estimate of the state variables in his control law, which may result in failure to meet the closed-loop specifications.