An Iterative Algorithm Combining Model Reduction and Control Design

A design strategy which integrates model reduction by modal cost analysis and a multi-objective controller design is proposed The necessary modeling and control algorithms are easily programmed in Matlab standard software. Hence, this method is very practical for controller design for large space structures. The design algorithm also presents a solution for the very important problem of "tuning" multiple loop controllers (MIMO). In stead of the single change that is used in standard Root Locus and gain and phase margin theories, the new method tunes multiple loop controllers from "low" to "high" gain in a systematic way in the design procedure. This design strategy is applied to NASA's Mini-Mast system at Langley Research Center.