Ship course stabilization by feedback linearization with adaptive object model

Abstract

The algorithm of ship course stabilization herein presented is based on a feedback linearization controller with adaptive object model. The described method, consisting in current approximation of unknown object model functions by neuro-fuzzy approximators, represents a new generation of adaptive control methods. The implementation of this algorithm, which may constitute an executive module of a navigational decision support system, will contribute to a higher degree of automation and navigational safety improvement.