Multivariable weighted LQG control design in the iterative Zang scheme

The combined frame of the process identification and the model based controller design requires the conformity between the control law and the identification criterion. This demand leads to an iterative scheme, namely the identification respects the control law´s attitude and the control design is based on the identified model. This paper presents an iterative model based control design method that extends the Zang scheme elaborated for single-input single-output (SISO) case to multi-input multi-output (MIMO) systems. Applying a weighting strategy this method results in a controller that matches the performance characteristic achievable with unknown plant although it has been designed on the basis of a reduced complexity identified model. Moreover, this paper shows an academic example for practical application of the theoretical results.