TY - GEN
T1 - Adaptive cubature strong tracking information filter using variational Bayesian method
AU - Ge, Quanbo
AU - Wen, Chenglin
AU - Chen, Shaodong
AU - Sun, Ruoyu
AU - Li, Yuan
N1 - Publisher Copyright:
© IFAC.
PY - 2014
Y1 - 2014
N2 - For most practical nonlinear state estimation problems, the conventional nonlinear filters do not usually work well for some cases, such as inaccurate system model, sudden change of state-interested and unknown variance of measurement noise. In this paper, an adaptive cubature strong tracking information filter using variational Bayesian (VB) method is proposed to cope with these complex cases. Firstly, the strong tracking filtering (STF) technology is used to integrally improve performance of cubature information filter (CIF) aiming at the first two cases and an iterative scheme is presented to effectively evaluate a strong tracking fading factor. Secondly, the VB method is introduced to iteratively evaluate the unknown variance of measurement noise. Finally, the novel adaptive cubature information filter is obtained by perfectly combining the STF technology with the VB method, where the variance estimation provided by the VB method guarantees normal running of the strong tracking functionality.
AB - For most practical nonlinear state estimation problems, the conventional nonlinear filters do not usually work well for some cases, such as inaccurate system model, sudden change of state-interested and unknown variance of measurement noise. In this paper, an adaptive cubature strong tracking information filter using variational Bayesian (VB) method is proposed to cope with these complex cases. Firstly, the strong tracking filtering (STF) technology is used to integrally improve performance of cubature information filter (CIF) aiming at the first two cases and an iterative scheme is presented to effectively evaluate a strong tracking fading factor. Secondly, the VB method is introduced to iteratively evaluate the unknown variance of measurement noise. Finally, the novel adaptive cubature information filter is obtained by perfectly combining the STF technology with the VB method, where the variance estimation provided by the VB method guarantees normal running of the strong tracking functionality.
KW - Cubature information filter
KW - Nonlinear system
KW - Strong tracking filtering
KW - Unknown variance of measurement noise
KW - Variational Bayesian
UR - http://www.scopus.com/inward/record.url?scp=84929774201&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84929774201&partnerID=8YFLogxK
U2 - 10.3182/20140824-6-za-1003.00558
DO - 10.3182/20140824-6-za-1003.00558
M3 - Conference contribution
AN - SCOPUS:84929774201
T3 - IFAC Proceedings Volumes (IFAC-PapersOnline)
SP - 5945
EP - 5950
BT - 19th IFAC World Congress IFAC 2014, Proceedings
A2 - Boje, Edward
A2 - Xia, Xiaohua
PB - IFAC Secretariat
T2 - 19th IFAC World Congress on International Federation of Automatic Control, IFAC 2014
Y2 - 24 August 2014 through 29 August 2014
ER -