Current Output Observer for Discrete-Time Nonlinear Stochastic Systems

Authors

Tongyan Zhai, Marquette University
Edwin E. Yaz, Marquette UniversityFollow
Chung Seop Jeong, Marquette University

Document Type

Article

Publication Date

2008

Publisher

Elsevier

Source Publication

IFAC Proceedings Volumes

Source ISSN

1474-6670

Original Item ID

DOI: 10.3182/20080706-5-KR-1001.01664

Abstract

A Current Output Observer is presented and its estimation error performance is compared to that of the Extended Kalman Filter. It is shown that performance improvement can be obtained by this new scheme with minor increase in computational load. In order to obtain stronger results, scalar nonlinear stochastic systems are focused on. These systems are categorized based on the derivatives of their nonlinear functions. It is shown that different state estimation performance is achieved when the Current Output Observer is applied to scalar nonlinear systems in these different categories, which allows the convergence property to be known before its implementation. This provides insight into what is going to happen in applications, e.g. for a nonlinear estimator used in chaotic synchronization. Simulation studies involving nonlinear estimation - based chaotic synchronization complement the theory presented.

Comments

IFAC Proceedings Volumes, Vol. 41, No. 2 (2008): 9834-9839. DOI.

Recommended Citation

Zhai, Tongyan; Yaz, Edwin E.; and Jeong, Chung Seop, "Current Output Observer for Discrete-Time Nonlinear Stochastic Systems" (2008). Electrical and Computer Engineering Faculty Research and Publications. 762.
https://epublications.marquette.edu/electric_fac/762