Adaptive Estimation and Dual Control with Applications to Defense Communications System Controls.

Abstract : Adaptive linear filters with the capability of tracking nonstationary noise statistics have been developed. These algorithms offer potentially improved solution methods for problems with significant nonlinearities in the system and measurement equations, high-dimensional problems, and problems in which the noise statistics are not exactly known. The application to a precision pointing and tracking problem for a telescope has been carried out to illustrate the potential benefits of an adaptive estimation algorithm. The simulation results show a sizable reduction in the pointing error compared to a non-adaptive filter. The previously developed dual control algorithm has been successfully applied to a missile guidance problem. Some further theoretical work on dual control has also been performed. It has been shown that the certainty equivalence property holds for a general class of problems if and only if the control has no dual effect.