Monograph and paper notices
Evaluation Practice · 1986
A very large volume of work on the use of fuzzy logic techniques and representations in the modelling and control of dynamic systems now exists. As yet, however, relatively little attention has been directed to the complementry subject of estimation. Fuzzy logic based controllers are particularly appealing for processes where a reliable standard mathematical model on which to base analysis and design is "difficult to obtain". In practice, however, the implementation of such a controller may also require an estimator due, for example, to the fact that a required signal cannot be physically measured. The central aim of this paper is to report some basic work on the use of fuzzy models as estimators in a general applications area-the tracking of an agile target based on uncertain measurements. Some very basic comparisons are also drawn between the fuzzy estimators proposed here and "standard" algorithms used in this general area. Some areas for further research are also briefly noted.