Incremental Learning of Fuzzy Basis Function Networks with a Modified Version of Vector Quantization

In this paper, an algorithm for data-driven incremental learning of fuzzy basis function networks is presented. A modified version of vector quantization is exploited for rule evolution and incremental learning of the rules' antecedent parts. Antecedent learning is connected in a stable manner with a recursive learning of rule consequent functions with linear parameters. The paper is concluded with an evaluation of the proposed algorithm on high-dimensional measurement data for engine test benches.