抄録
Recently, it has been shown that the methods using PDLMI (Parameter-Dependent LMI) are effective for the robust H∞ performance analysis of the linear polytopic-type uncertain systems. Especially, the extended LMI approaches can avoid the product of the Lyapunov matrix and the system matrix by enlarging the size of LMI, and a less conservative analysis can be achieved by using these approaches. On the other hand, the redundant descriptor approaches are known as methods of consequentially enlarging the size of LMI by expanding the rep-resentation of the system. In this paper, we derive extended LMI conditions for the robust H∞ performance analysis from the standpoint of a redundant descriptor approach. Moreover, it is examined to decrease the number of concomitant variables of the extended LMI in order to shorten time to compute the solution.
