CONTROLLER REDUCTION BY COVARIANCE EQUIVALENT REALIZATIONS.
Conference Paper
Overview
Additional Document Info
View All
Overview
abstract
This paper presents a simple, non-iterative algorithm that uses a suboptimal projection approach to design reduced order controllers. This algorithm is based upon the theory of covariance equivalent realizations. Properties of reduced controllers, such as invariance, stability, Markov parameters, and minimal order, are presented. This algorithm, which is capable of extracting optimal, minimal LQG-controllers, is also shown to yield controllers that are equivalent to those obtained by component cost analysis.