Searching for basic properties obtaining robust implication operators in fuzzy control (Q1973341): Difference between revisions

In rule-based modeling and control, a design problem of choosing a suitable implication in constructing an inference mechanism is of paramount importance. A general design criterion to be met is the one about a ``robust'' behavior of the implication in the sense that such an implication should exhibit a good reasonable performance in a broad class of applications and in conjunction with different defuzzification methods. The study concentrates on two main classes of implication operators, namely (a) those being an extension of Boolean implications and (b) the implications associated with the extended Boolean conjunctions (and realized through \(t\)-norms and force implications). The experimental part illustrates various implications in the setting of fuzzy rule-based modeling. Finally, the paper attempts to set up some general selection criteria for the robust implication operators.