Discrete-time markovian-jump linear quadratic optimal control (Q3720440)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3950350
Language Label Description Also known as
default for all languages
No label defined
    English
    Discrete-time markovian-jump linear quadratic optimal control
    scientific article; zbMATH DE number 3950350

      Statements

      Discrete-time markovian-jump linear quadratic optimal control (English)
      0 references
      0 references
      0 references
      0 references
      1986
      0 references
      discrete-time jump linear systems
      0 references
      finite-state Markov chain
      0 references
      optimal control law
      0 references
      Discrete-time jump linear systems \(X_{K+1}=A_ K(r_ K)X_ K+B_ K(r_ K)u_ K\), \(K=K_ 0,...,N\), \(P\{r_{K+1}=j/r_ K=i\}=P_{K+1}(i,j)\) with initial state \(X(K_ 0)=X_ 0\), \(r(K_ 0)=r_ 0\) are considered. It is assumed that the x-process, and the control vector u are m-dimensional and that the form process \(\{r| K=K_ 0,...,N\}\) is a finite-state Markov chain taking values in \({\mathcal M}=\{1,2,...,M\}\). Further, it is assumed that the cost criterion is quadratic. First, the optimal control law is presented. This optimal control law is linear in \(X_ K\) at each time K, and it is different (in general) for each possible set of parameter values. Further, necessary and sufficient conditions for the existence of a steady-state optimal controller are given. The results are illustrated by examples.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references