State bounding with ellipsoidal set description of the uncertainty

Abstract

The paper presents recursive algorithms for state bounding, using ellipsoidal sets to describe the state uncertainties and to bound the process and observation noises. The algorithms optimize each stage of updating, according to the minimum-volume and minimum-trace criteria. Simulations compare the performance of the bounding algorithms with that of a Kalman filter, and investigate the influence of noise distribution within the bounds.