Open AccessRobust measure of non‐linearity‐based cubature Kalman filter
Author(s) -
Zhang Lei,
Li Sheng,
Zhang Enze,
Chen Qingwei
Publication year - 2017
Publication title -
iet science, measurement and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.418
H-Index - 49
eISSN - 1751-8830
pISSN - 1751-8822
DOI - 10.1049/iet-smt.2017.0108
Subject(s) - kalman filter , computation , linearity , filter (signal processing) , computer science , gaussian , measure (data warehouse) , algorithm , control theory (sociology) , noise (video) , degree (music) , gaussian noise , computational complexity theory , mathematics , artificial intelligence , data mining , computer vision , electronic engineering , engineering , physics , control (management) , quantum mechanics , acoustics , image (mathematics)
In this study, a novel robust measure of non‐linearity‐based cubature Kalman filter (RMoNCKF) is proposed to obtain good performance with lower computational burden. The proposed filter inherits the virtues of high accuracy of the high‐degree filter and computation efficiency of the low‐degree one. When the measure of non‐linearity (MoN) is evaluated and compared with the threshold in the dynamic system, the cubature rules nested in the RMoNCKF can be switched autonomously to decrease the computation complexity in the low non‐linear condition. Furthermore, the robust estimation technology can help to improve the value of MoN for the non‐Gaussian distributed case. Simulation results of target tracking and integrated navigation system demonstrate that the RMoNCKF can have a close performance to the fifth‐degree CKF with less computation time. In the circumstances of the time‐varying noise and contaminated Gaussian distributed noise, the RMoNCKF outperforms the UKF, the third‐degree CKF and fifth‐degree CKF.