Abstract
In general, fuzzy modeling requires two stages: structure identification (generating the fuzzy rule base) and parameter learning (optimizing parameters in fuzzy rules). Here, we present an on-line algorithm for competitive learning and optimization of fuzzy models. Differing from existing methods, in this approach the structure identification and parameter optimization of the fuzzy model can be carried out automatically, using on-line acquisition of data. We demonstrate this approach by applying it to different types of nonlinear system modeling.