Iterative Kalman Filter and Related Algorithms for Non-linear Systems

Abstract

This paper proposes a comparative analysis of different state estimation techniques on linear and non-linear systems. Estimation is the process of finding a value that is usable even if the subject of interest is uncertain, delayed or corrupted due to noise. An Iterative Kalman Filter has been developed for a class of uncertain discrete-time system with delay. It extends, the KF and EKF to the case in which the underlying system is subjected to norm-bounded uncertainties and constant state delay. The IKF is a robust version of KF, but with the necessary modification to account for the parameter uncertainty as well as delay.