Optimized distributed fusion filtering for singular systems with fading measurements and stochastic nonlinearity

AIMS Mathematics

Jun Hu¹  Hui Yu²  Dongyan Chen²  Chen Wang² 

[1] 1. Department of Mathematics, Harbin University of Science and Technology, Harbin 150080, China 2. School of Automation, Harbin University of Science and Technology, Harbin 150080, China;1. Department of Mathematics, Harbin University of Science and Technology, Harbin 150080, China;

关键词: singular systems;    distributed fusion filter;    innovation analysis approach;    singular value decomposition;    multi-sensors;   

DOI  :  10.3934/math.2022143

来源: DOAJ

【 摘 要 】

In this paper, the problem of optimized distributed fusion filtering is considered for a class of multi-sensor singular systems in the presence of fading measurements and stochastic nonlinearity. By utilizing the standard singular value decomposition, the multi-sensor stochastic singular systems are simplified to two reduced-order nonsingular subsystems (RONSs). The local filters (LFs) with corresponding error covariance matrices are proposed for RONSs via the innovation analysis approach. Then, on the basis of the matrix-weighted fusion estimation algorithm, the distributed fusion filters (DFFs) are designed for RONSs with multiple sensors in the linear minimum variance sense. Moreover, the DFFs are obtained by utilizing the state transformation for original singular systems. It can be observed that the DFFs have better accuracy in contrast with the LFs. Finally, an illustrate example is put forward to verify the feasibility of the proposed fusion filtering scheme.

【 授权许可】

Unknown