Averaged dynamics and pole assignment for a class of stochastic systems

From MaRDI portal





The pth moment Lyapunov exponent (p a natural number) of the linear control system \(\dot x{}_ t=A(r_ t)x_ t+B(r_ t)u_ t\), \(r_ t\) an N-state Markov process with transition rates \(\Pi =(\pi_{ij})\), is represented as dominant eigenvalue of a matrix \(E_ p(A,\Pi)\) associated with a corresponding averaged system. For feedback control \(u=\Gamma (r_ t)x\) a prescribed spectrum is given to \(F_ p(\tilde A,\Pi)\), \(\tilde A=A+B\Gamma\), by assigning suitable spectra to \(F_ p(\tilde A,\alpha \Pi)\), \(0\leq \alpha \leq 1\), starting with the eigenvalues of the \(\tilde A(i)\) for \(F_ p(\tilde A,0)=diag(\tilde A(1),...,\tilde A(N))\) when \(\alpha =0\) and moving continuously to \(\alpha =1\) (homotopy algorithm).











This page was built for publication: Averaged dynamics and pole assignment for a class of stochastic systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3812149)