DocumentCode :
819781
Title :
Sequential estimation for continuous-time linear dynamic systems with completely unknown inputs and duality with singular control
Author :
Mehra, R.K.
Author_Institution :
Harvard University, Cambridge, MA, USA
Volume :
20
Issue :
5
fYear :
1975
fDate :
10/1/1975 12:00:00 AM
Firstpage :
675
Lastpage :
677
Abstract :
Completely unknown but deterministic (or Fisher-nnknown) inputs often occur in practice and are approximated by Gauss-Markov processes in applying Kalman-Bucy filtering results. In this paper, the exact filtering equations for such inputs are presented in terms of an information-form filter. The equations are obtained by a limit procedure from the corresponding equations for discrete-time systems. Using duality arguments, the necessary and sufficient conditions for singular control problems are shown to follow from these results.
Keywords :
Linear systems, time-varying continuous-time; Sequential estimation; Singular optimal control; State estimation; Bayesian methods; Control systems; Equations; Information filtering; Information filters; Maximum likelihood estimation; Nonlinear filters; Optimal control; Smoothing methods; State estimation;
fLanguage :
English
Journal_Title :
Automatic Control, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9286
Type :
jour
DOI :
10.1109/TAC.1975.1101064
Filename :
1101064
Link To Document :
بازگشت