Open AccessA Novel Step‐by‐Step High‐Order Extended Kalman Filter Design for a Class of Complex Systems with Multiple Basic Multipliers
Author(s) -
Xiaohui Sun,
Chenglin Wen,
Tao Wen
Publication year - 2021
Publication title -
chinese journal of electronics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.267
H-Index - 25
eISSN - 2075-5597
pISSN - 1022-4653
DOI - 10.1049/cje.2021.02.005
Subject(s) - kalman filter , latent variable , state variable , extended kalman filter , linearization , latent class model , control theory (sociology) , invariant extended kalman filter , mathematics , alpha beta filter , nonlinear system , computer science , variable (mathematics) , algorithm , mathematical optimization , artificial intelligence , statistics , mathematical analysis , physics , control (management) , quantum mechanics , thermodynamics , moving horizon estimation
A novel step‐by‐step linearization high‐order Extended Kalman filter SH‐EKF is designed for a class of nonlinear systems composed of linear functions and the product of several separable basic functions. The basic functions in the state and measurement models are defined as latent variables; the state and measurement models are equivalently formulated into pseudo‐linear models based on the combination of the original variable and the latent variables; latent variables are regarded as new variables, and a dynamic linear model between each latent variable and other latent variables with original state is established; the measurement model is rewritten into the first‐order linear product form between the current state and each latent variable; latent variables are solved by Kalman filter step by step, and a stepwise linearized high‐order extended Kalman filter is designed. Illustration examples are presented to demonstrate the effectiveness of the new algorithm.