Abstract
This paper discusses a linear feedback control system subjected to gaussian white noise. and gives an optimal linear feedback control law which minimizes the integrated-mean-square error criterion. The method used in this paper is based upon the analysis of Ricatti-type matrix equations derived from the optimal regulator problem. The duality relating covariance matrix equation and mean-square deviation equation plays a central role in the evaluation of the rms error.
The main result is that the optimal linear feedback control law minimizing the rms criterion is equivalent to the feedback law of the optimal regulator.
