Interval State Estimator Design for Linear Parameter Varying (LPV) Systems

Abstract

This brief proposes a new method to develop interval state estimator for a class of LPV systems subject to external disturbances and uncertainties. Usually, the design of such estimators is based on the system cooperativity. The proposed technique uses the previous input-output values and the observability matrix to develop an interval state estimator that can provide extra degree of freedom. The main advantage of the technique compared to the standard interval observer design method for cooperative and positive systems is that it does not require the cooperativity of the system dynamics. It is shown that the convergence time of the interval width is finite and hence proved that estimation errors are bounded. The interval state observation error bounds are calculated a-priori for the given bounded disturbances. A numerical example illustrates the high potential of the proposed technique.