Innovation-Modified Kalman Filtering for A Class of Stochastic Nonlinear Systems

This paper is concerned with the innovation-detected Kalman filtering problem for a class of stochastic nonlinear systems. Owing to the temporal failure of sensors caused by undesirable environmental circumstances, the measurement outputs of sensors might suffer from unexpected outliers. In this case, a detection scheme of the sensor measurement is employed to alleviate its unfavorable effect on the filtering performance. With the aid of the stochastic analysis techniques, a sufficient condition is established to guarantee that there exists an upper bound of the filtering error variance. Then, the filter gain matrix is determined by minimizing such an upper bound. A simulation example is finally presented to illustrate the validity of the proposed filtering scheme.