Citation

BibTex format

@article{Ge,
author = {Ge, M and Kerrigan, EC},
journal = {Automatica},
title = {Noise Covariance Identification for Nonlinear Systems using Expectation Maximization and Moving Horizon Estimation},
url = {http://hdl.handle.net/10044/1/41282},
}

RIS format (EndNote, RefMan)

TY  - JOUR
AB - In order to estimate states from a noise-driven state space system, the state estimator requires a priori knowledge of bothprocess and output noise covariances. Unfortunately, noise statistics are usually unknown and have to be determined fromoutput measurements. Current expectation maximization (EM) based algorithms for estimating noise covariances for nonlinearsystems assume the number of additive process and output noise signals are the same as the number of states and outputs,respectively. However, in some applications, the number of additive process noises could be less than the number of states. Inthis paper, a more general nonlinear system is considered by allowing the number of process and output noises to be smaller orequal to the number of states and outputs, respectively. In order to estimate noise covariances, a semi-definite programmingsolver is applied, since an analytical solution is no longer easy to obtain. The expectation step in current EM algorithms relyon state estimates from the extended Kalman filter (EKF) or smoother. However, the instability and divergence problems ofthe EKF could cause the EM algorithm to converge to a local optimum that is far away from true values. We use movinghorizon estimation instead of the EKF/smoother so that the accuracy of the covariance estimation in nonlinear systems canbe significantly improved.
AU - Ge,M
AU - Kerrigan,EC
SN - 0005-1098
TI - Noise Covariance Identification for Nonlinear Systems using Expectation Maximization and Moving Horizon Estimation
T2 - Automatica
UR - http://hdl.handle.net/10044/1/41282
ER -