Fuzzy Kalman Filter using Linear Matrix Inequalities

Abstract

The Kalman filter has been extensively used in different applications due to its strengths in estimating the system states under noisy observations. In this paper, a modification of the classical Kalman filter for nonlinear state estimation is presented; firstly, a polytopic set of linear discrete-time models based on a Takagi-Sugeno inference system is used to describe the nonlinear operating region. The stabilizing gains of the linear filters are calculated using Linear Matrix Inequalities (LMI), the proposal is evaluated through simulations.