Set-Membership Filtering Subject to Impulsive Measurement Outliers: A Recursive Algorithm

This paper is concerned with the set-membership filtering problem for a class of linear time-varying systems norm-bounded noises and impulsive measurement outliers. A new representation proposed to model outlier by an signal whose minimum interval length (i.e., duration between two adjacent signals) norm norms all are larger than certain thresholds that adjustable according engineering practice. In order guarantee satisfactory performance, so-called parameter-dependent filter put forward capable generating ellipsoidal region containing true system state. First, novel detection strategy developed, based on dedicatedly constructed input-output model, examine whether received corrupted outlier. Then, through outcome detection, gain matrix desired corresponding calculated solving recursive difference equations. Furthermore, ultimate boundedness issue thoroughly investigated. Finally, simulation example provided demonstrate effectiveness our strategy.