Local mode dependent output feedback control of uncertain Markovian jump large-scale systems

This paper is concerned with the output feedback guaranteed cost control problem for a class of uncertain stochastic large-scale systems governed by a random parameter. The uncertainties are assumed to satisfy integral quadratic constraints, and the random parameter is a Markov process. A sufficient condition is established for the design of decentralized output feedback guaranteed cost controllers which use local system states and local system operation modes to produce local control inputs, and ensure suboptimal global quadratic performance. The condition is given in terms of a set of rank constrained linear matrix inequalities. A numerical example and simulations are also provided to illustrate the theory.