Recursive filtering for complex networks using non‐linearly coupled UKF. Issue 4 (9th January 2018)

• Record Type: Journal Article
• Authors: Li, Wenling; Meng, Cong; Jia, Yingmin; Du, Junping
• Is Part Of: IET control theory & applications. Volume 12:Issue 4(2018)
• Page Start: 549
• Page End: 555
• Publication Date: 2018-01-09
• Subjects: Kalman filters -- nonlinear filters -- stochastic processes -- recursive estimation -- target tracking; recursive filtering problem -- discrete‐time complex networks -- nonlinearly coupled UKF -- nonlinearly coupled terms -- coupled unscented Kalman filter -- sigma points -- stochastic analysis technique -- estimation errors boundedness -- upper bound -- coupling strength -- multiple interacting target tracking; Control theory -- Periodicals; Automatic control -- Periodicals; 629.8312
• DOI: 10.1049/iet-cta.2017.0738
• Languages: English
• ISSNs: 1751-8644

Abstract

Abstract : This study is concerned with the recursive filtering problem for a class of discrete‐time complex networks with non‐linearly coupled terms. A coupled unscented Kalman filter (UKF) is developed, where the sigma points of the UKF are propagated by introducing the coupled terms. By using the stochastic analysis technique, a sufficient condition is established to guarantee the boundedness of the estimation errors, where an upper bound for the coupling strength is derived. It is shown that the upper bound of the coupling strength is inversely proportional to the number of nodes. A numerical example involving tracking of multiple interacting targets is provided to verify the effectiveness of the proposed filter.