New developments in state estimation for nonlinear systems

M. Nørgaard

doi:10.1016/S0005-1098(00)00089-3

2000-11-01

849

PRE

M. Nørgaard *

Niels Kjølstad Poulsen

Ole Ravn

DOI：10.1016/S0005-1098(00)00089-3

原文链接

原文求助

摘要

En 中文

State estimators for nonlinear systems are derived based on polynomial approximations obtained with a multi-dimensional interpolation formula. It is shown that under certain assumptions the estimators perform better than estimators based on Taylor approximations. Nevertheless, the implementation is significantly simpler as no derivatives are required. Thus, it is believed that the new state estimators can replace well-known estimators, such as the extended Kalman filter (EKF) and its higher-order relatives, in most practical applications. (C) 2000 Elsevier Science Ltd. All rights reserved.

Keyword:

state estimation

nonlinear filters

interpolation algorithms

multivariable polynomials

extended Kalman filters

parameter estimation