An Adaptive Formulation of the Sliding Innovation Filter

Abstract

In this paper, an adaptive formulation of the sliding innovation filter (SIF) is presented. The SIF is a recently proposed estimation strategy that has demonstrated robustness to modeling errors and uncertainties. It utilizes a switching gain that is a function of the innovation (measurement error) and sliding boundary layer term. In this paper, a time-varying sliding boundary layer is derived based on minimizing the state error covariance. The resulting solution creates an adaptive formulation of the SIF. The adaptive SIF is applied on a linear aerospace system, and is compared with the well-known Kalman filter (KF) and the standard SIF. The results demonstrate the robustness of the new estimation strategy in the presence of modeling uncertainties and system faults.