Fault Estimation for A Class of Descriptor Systems - An Adaptive Robust Extended Kalman Filter Approach

This paper proposes an adaptive Robust Extended Klaman filter for a class of non-linear descriptor systems with unknown system noise. Firstly, a robust bound is given to decrease the influence of the linearization error on the estimation accuracy; an adaptive algorithm is introduced to implement an unbiased estimation of the noise, then; an numeral example is given to show the effectiveness of the method at last.