Open AccessThe optimal estimation of parameters of models of controlled stochastic systems based on the experiment design
Author(s) -
В. И. Денисов,
Vladimir M. Chubich,
Elena V. Filippova
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1333/3/032020
Subject(s) - control theory (sociology) , covariance , identification (biology) , noise (video) , covariance matrix , computer science , system identification , state (computer science) , estimation , estimation theory , software , mathematics , mathematical optimization , control (management) , algorithm , engineering , statistics , artificial intelligence , data modeling , botany , systems engineering , database , image (mathematics) , biology , programming language
The procedure of active identification, which is resistant to the appearance of anomalous observations, includes robust estimation of unknown parameters and optimal design of input signals for models of non-stationary linear discrete systems is proposed. The general case of the entry of unknown parameters into the equations of state and observation, the initial condition and the covariance noise matrix of the system is considered. Using the developed software, the efficiency of this procedure is demonstrated by the example of a direct current motor control system.