The Kleinman iteration for nonstabilizable systems (Q1402892)

From MaRDI portal
scientific article
Language Label Description Also known as
English
The Kleinman iteration for nonstabilizable systems
scientific article

    Statements

    The Kleinman iteration for nonstabilizable systems (English)
    0 references
    0 references
    0 references
    0 references
    31 August 2003
    0 references
    A standard approach to obtaining a positive semidefinite solution (if it exists) of the algebraic matrix Riccati equation (ARE) \(A^*P+PA-PBB^*P+C^*C=0\) is by using the Kleinman iteration procedure. The procedure consists of solving consecutively a sequence of Lyapunov equations \(A^*_iP_{i+1} +P_{i+1}A_i+ C^*_iC_i=0\), \(i=0,1,\dots\), starting with a suitable initial matrix \(P_0\), and represents essentially an application of Newton's method to (ARE). So far in the literature, the stabilizability of \((A,B)\) was assumed to guarantee that the Kleinman iteration converges. In the reviewed paper, the stabilizability hypothesis is disposed of, at the expense of assuming a certain property of the initial matrix \(P_0\). As a result, a new criterion for the existence of positive semidefinite solutions of (ARE) is proved.
    0 references
    0 references
    0 references
    0 references
    0 references
    algebraic Riccati equation
    0 references
    Newton's method
    0 references
    Kleinman iteration
    0 references
    Lyapunov equation
    0 references
    continuous-time linear systems
    0 references
    nonstabilizable systems
    0 references
    linear-quadratic optimal control
    0 references
    stabilizability
    0 references
    0 references
    0 references
    0 references