Dynamic feedback control of stochastic-parameter systems

From MaRDI portal





The author considers the linear stochastic-parameter discrete-time system (1) \(x_{k+1}=A_ k(\omega)x_ k+B_ k(\omega)u_ k+v_ k\), \(y_ k=C_ k(\omega)x_ k+w_ k\), where \(x\in R^ n\) is the state, \(u_ k\in R^ m\) is the input (control), \(y_ k\in R^ p\) is the output, \(v_ k\in R^ n\) and \(w_ k\in R^ p\) are the input and output additive zero-mean white noise sequences which may be correlated with each other, \(x_ 0\) is deterministic but unknown. The matrices \(A_ k(\omega)\), \(B_ k(\omega)\) and \(C_ k(\omega)\) have appropriate dimensions and have elements which are purely random sequences with constant and bounded moments. To stabilize the system (1) the author proposes linear feedback \(u_ k=K^ c\hat x_ k\) of the state estimate \(x_ k\), which is found by the observer \(\hat x_{k+1}=\bar A_ k\hat x_ k+\bar B_ ku_ k+K^ 0(y_ k-\bar C\hat x_ k)\) where the overbar indicates the expectation operation, \(K^ c\), \(K^ 0\) are respectively the constant controller and observer gain matrices. The author obtains sufficient conditions for stabilizability which depend both on the first and second moments of the stochastic parameters. The obtained results are illustrated by a numerical example.




Cited in
(19)








This page was built for publication: Dynamic feedback control of stochastic-parameter systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3799929)