A comparative study of neural network structures in identification of nonlinear systems

Authors
Efe, MO Kaynak, O
Citation
Mo. Efe et O. Kaynak, A comparative study of neural network structures in identification of nonlinear systems, MECHATRONIC, 9(3), 1999, pp. 287-300
Citations number
8
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
MECHATRONICS
ISSN journal
09574158 → ACNP
Volume
9
Issue
3
Year of publication
1999
Pages
287 - 300
Database
ISI
SICI code
0957-4158(199904)9:3<287:ACSONN>2.0.ZU;2-Z
Abstract
This paper investigates the identification of nonlinear systems by neural n etworks. As the identification methods, Feedforward Neural Networks (FNN), Radial Basis Function Neural Networks (RBFNN), Runge-Kutta Neural Networks (RKNN) and Adaptive Neuro Fuzzy Inference Systems (ANFIS) based identificat ion mechanisms are studied and their performances are comparatively evaluat ed on a three degrees of freedom anthropomorphic robotic manipulator. (C) 1 998 Elsevier Science Ltd. All rights reserved.