A Rapid Learning Approach for Developing Fuzzy Controllers

Abstract In this paper we propose an approach for rapid learning an important type of fuzzy controllers To specify linguistic terms the B spline basis functions are used for input variables and fuzzy singletons for output variables Product is chosen as the fuzzy conjunction and centroid as the defuzzi cation method By appropriately designing the rule base a fuzzy logic controller can be interpreted as a B spline interpolator Such a fuzzy controller can learn to approximate any known data sequences and to minimise a certain cost functions By choosing a suitable cost function the learning process can converge rapidly We present some applications of this approach in supervised learning especially for function approximation The approach can also be extended to the problem of unsupervised learning