Application of Matrix Minimum Principle to a linear decentralized optimal control in descriptor systems is studied in this thesis. Linear-quadratic index of performance with Gaussian initial state is considered. The necessary and sufficient conditions for optimality are derived
An additional constraint is imposed such that the controllers are linear function of output y(t) rather than of the state vector x(t). The optimal gain matrix Gi* is then specified by the necessary conditions.
Two examples are developed to illustrate the concept.
If you have any questions please contact the ETD Team, libetd@njit.edu.