TY - JOUR
T1 - A unified framework for M-estimation based robust Kalman smoothing
AU - Wang, Hongwei
AU - Li, Hongbin
AU - Zhang, Wei
AU - Zuo, Junyi
AU - Wang, Heping
N1 - Publisher Copyright:
© 2018 Elsevier B.V.
PY - 2019/5
Y1 - 2019/5
N2 - We consider the robust smoothing problem for a state-space model with outliers in measurements. A unified framework for robust smoothing based on M-estimation is developed, in which the robust smoothing problem is formulated by replacing the quadratic loss for measurement fitting in the conventional Kalman smoother by a robust cost function from robust statistics. The majorization-minimization method is employed to iteratively solve the formulated robust smoothing problem. In each iteration, a surrogate function is constructed for the robust cost, which enables the states update procedure to be implemented in a similar way as that in a conventional Kalman smoother with a reweighted measurement covariance. Numerical experiments show that the proposed robust approach outperforms the traditional Kalman smoother and several robust filtering methods.
AB - We consider the robust smoothing problem for a state-space model with outliers in measurements. A unified framework for robust smoothing based on M-estimation is developed, in which the robust smoothing problem is formulated by replacing the quadratic loss for measurement fitting in the conventional Kalman smoother by a robust cost function from robust statistics. The majorization-minimization method is employed to iteratively solve the formulated robust smoothing problem. In each iteration, a surrogate function is constructed for the robust cost, which enables the states update procedure to be implemented in a similar way as that in a conventional Kalman smoother with a reweighted measurement covariance. Numerical experiments show that the proposed robust approach outperforms the traditional Kalman smoother and several robust filtering methods.
KW - M-estimation
KW - Majorization-minimization
KW - Robust Kalman smoother
KW - State-space modeling
UR - http://www.scopus.com/inward/record.url?scp=85059319216&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85059319216&partnerID=8YFLogxK
U2 - 10.1016/j.sigpro.2018.12.017
DO - 10.1016/j.sigpro.2018.12.017
M3 - Article
AN - SCOPUS:85059319216
SN - 0165-1684
VL - 158
SP - 61
EP - 65
JO - Signal Processing
JF - Signal Processing
ER -