Interval-valued fuzzy controller based on verbal model of object (Q1114638)

This paper presents further development of the idea to construct control algorithms on the basis of verbal models of the controlled objects. This is achieved using the so-called interval-valued fuzzy method of approximate inference. The above approach has some advantages: it is possible to perform prediction and to generate control actions according to the defined performance index, which is not always possible in the case of the approach based on the formalisation of the process operator strategy. An outline of the main idea is illustrated by means of a simple example. Next, the author provides remarks concerning the generalization to the case of complex objects. This method enables a more adequate fuzzy mapping of the available information about the object, and it is constructed in a way which allows its effective processing in a computer.