Adaptive Kalman Filtering through Fuzzy Logic

− In this paper a development of an adaptive Kalman filter through a fuzzy inference system (FIS) is outlined. The adaptation is concerned with the imposition of conditions under which the filter measurement noise covariance matrix R or the process noise covariance matrix Q are estimated. The adaptive adjustment is carried out using a FIS based on the whiteness of the filter innovation sequence (IS) and employing the covariance-matching technique. If a statistical analysis of the IS shows discrepancies with its expected statistics then the FIS adjusts a factor through which the matrices R or Q are estimated. This fuzzy adaptive Kalman filter is tested on a numerical example. The results are compared with these obtained using a conventional Kalman filter and a traditionally adapted Kalman filter. The fuzzy-adapted Kalman filter showed better results than its traditional counterparts.