Successive D.C. relaxation of parametrized LMIs - control applications

This paper presents a global optimization algorithm using difference of multiconvex relaxation successively for parametrized LMI problems. The problem is formulated as a minimax problem. On branch and bound, the relaxation is performed iteratively to have a feasible solution of the parametrized LMI problem. Numerical examples for stability analysis and L/sub 2/ gain performance analysis of linear parameter varying systems are shown.