A recursive algorithm for identification and control of large-scale systems

The paper presents new recursive algorithms for identification and control of linear time invariant discrete time systems of large size. A variant of the observable canonical form (Weinert and Anton 1972) of the state model is chosen for developing solution of the identification problem. The proposed solution turns out to be computationally much simpler than the existing solutions. It is also shown that centralized and decentralized output feedback controllers can be found by a simple extension of the identification algorithms. The results are illustrated for a system obtained by interconnecting three smaller subsystems.