WIT Press

Neuro-fuzzy Networks With Hybrid Learning

Price

Free (open access)

Paper DOI

10.2495/AI970121

Volume

19

Pages

16

Published

1997

Size

107 kb

Author(s)

Hamid Beigy & Ali Mohammadzadeh Eydgahi

Abstract

In this paper, two neuro-fuzzy networks based on fuzzy logic and elastic fuzzy logic with their learning algorithms have been proposed. The learning process is based on the back propagation algorithm. To speed-up the learning process three learning algorithms using a momentum term are developed. An example of two link planar robot is used to show usefulness of the proposed networks and their learning algorithms. 1 Introduction Fuzzy logic systems could be considered as one of the best tool for modeling the human reasoning that is approximate and partial in essence. Neural networks are computational paradigm which are suitable to handle partial and approximate reasoning [1]. There are increasing interests in

Keywords