Parametric robust H2 control design using iterative linear matrix inequalities synthesis

The numerical performance of a new linear matrix inequalities (LMI)-based design technique for robust H2-controller was analyzed for systems with real parametric uncertainty. This technique iterates between solving LMIs for the analysis parameters with the controller values fixed and LMIs for the controller values with the multiplier parameters fixed. A potential advantage of the LMI approach is the low implementation overhead associated with the optimization, especially given the simplicity of current semidefinite programming interfaces. This advantage also simplifies the extension of the algorithm to include more general stability multipliers.