Ivan G. Ivanov, Nonka Georgieva
A Linear Matrix Inequality Method for Stochastic Model on Markov Jump Linear Systems
Summary:
We consider a special class of linear quadratic stochastic models on Markov jump linear system. The aim is to find the best control function for a model. The search of the control function passes trough the computation of the maximal solution to a system of the general discrete time Riccati equations. An effective method for finding the maximal solution is the method of a linear matrix inequality (a standard method). In this paper we present two new modifications of the method of a linear matrix inequality. The numerical experiments for comparing the computational characteristics of the modified methods and the standard method are executed. The numerical experiments show the effectiveness of new methods to the standard method.
