Self tuning of computed torque gains by using neural networks with flexible structures

M. Teshnehlab, Keigo Watanabe

Research output: Contribution to journalArticle

29 Citations (Scopus)

Abstract

The paper principally describes the design of an artificial neural network using flexible sigmoid unit functions (FSUFs), referred to as flexible sigmoid function networks (FSFNs), to achieve both a high flexibility and a high learning ability in neural network structures from a given set of teaching patterns. An FSFN can generate an appropriate shape of the sigmoid function for each of the individual hidden- and output-layer units, in accordance with the specified inputs, desired output(s) and applied system. The paper proposes a learning method in which not only connection weights but also the sigmoid functions may be adjusted. The learning algorithm is derived by using the well known back-propagation algorithm. To demonstrate the validity of the proposed method, we apply the FSFN to the construction of a self-tuning computed torque controller for a two-link manipulator. It is then shown that the controller based on the FSFN gives a better control performance than that based on the traditional neural network.

Original languageEnglish
Pages (from-to)235-242
Number of pages8
JournalIEE Proceedings: Control Theory and Applications
Volume141
Issue number4
DOIs
Publication statusPublished - Jul 1994
Externally publishedYes

Fingerprint

Flexible structures
torque
Torque
Tuning
tuning
Neural networks
learning
controllers
Controllers
Backpropagation algorithms
output
Learning algorithms
Manipulators
manipulators
flexibility
Teaching
education

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering
  • Instrumentation

Cite this

Self tuning of computed torque gains by using neural networks with flexible structures. / Teshnehlab, M.; Watanabe, Keigo.

In: IEE Proceedings: Control Theory and Applications, Vol. 141, No. 4, 07.1994, p. 235-242.

Research output: Contribution to journalArticle

@article{6eec65daf0f74e718f23c4a8f42a5436,
title = "Self tuning of computed torque gains by using neural networks with flexible structures",
abstract = "The paper principally describes the design of an artificial neural network using flexible sigmoid unit functions (FSUFs), referred to as flexible sigmoid function networks (FSFNs), to achieve both a high flexibility and a high learning ability in neural network structures from a given set of teaching patterns. An FSFN can generate an appropriate shape of the sigmoid function for each of the individual hidden- and output-layer units, in accordance with the specified inputs, desired output(s) and applied system. The paper proposes a learning method in which not only connection weights but also the sigmoid functions may be adjusted. The learning algorithm is derived by using the well known back-propagation algorithm. To demonstrate the validity of the proposed method, we apply the FSFN to the construction of a self-tuning computed torque controller for a two-link manipulator. It is then shown that the controller based on the FSFN gives a better control performance than that based on the traditional neural network.",
author = "M. Teshnehlab and Keigo Watanabe",
year = "1994",
month = "7",
doi = "10.1049/ip-cta:19941225",
language = "English",
volume = "141",
pages = "235--242",
journal = "IEE Proceedings: Control Theory and Applications",
issn = "1350-2379",
publisher = "Institute of Electrical Engineers",
number = "4",

}

TY - JOUR

T1 - Self tuning of computed torque gains by using neural networks with flexible structures

AU - Teshnehlab, M.

AU - Watanabe, Keigo

PY - 1994/7

Y1 - 1994/7

N2 - The paper principally describes the design of an artificial neural network using flexible sigmoid unit functions (FSUFs), referred to as flexible sigmoid function networks (FSFNs), to achieve both a high flexibility and a high learning ability in neural network structures from a given set of teaching patterns. An FSFN can generate an appropriate shape of the sigmoid function for each of the individual hidden- and output-layer units, in accordance with the specified inputs, desired output(s) and applied system. The paper proposes a learning method in which not only connection weights but also the sigmoid functions may be adjusted. The learning algorithm is derived by using the well known back-propagation algorithm. To demonstrate the validity of the proposed method, we apply the FSFN to the construction of a self-tuning computed torque controller for a two-link manipulator. It is then shown that the controller based on the FSFN gives a better control performance than that based on the traditional neural network.

AB - The paper principally describes the design of an artificial neural network using flexible sigmoid unit functions (FSUFs), referred to as flexible sigmoid function networks (FSFNs), to achieve both a high flexibility and a high learning ability in neural network structures from a given set of teaching patterns. An FSFN can generate an appropriate shape of the sigmoid function for each of the individual hidden- and output-layer units, in accordance with the specified inputs, desired output(s) and applied system. The paper proposes a learning method in which not only connection weights but also the sigmoid functions may be adjusted. The learning algorithm is derived by using the well known back-propagation algorithm. To demonstrate the validity of the proposed method, we apply the FSFN to the construction of a self-tuning computed torque controller for a two-link manipulator. It is then shown that the controller based on the FSFN gives a better control performance than that based on the traditional neural network.

UR - http://www.scopus.com/inward/record.url?scp=0028463939&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028463939&partnerID=8YFLogxK

U2 - 10.1049/ip-cta:19941225

DO - 10.1049/ip-cta:19941225

M3 - Article

AN - SCOPUS:0028463939

VL - 141

SP - 235

EP - 242

JO - IEE Proceedings: Control Theory and Applications

JF - IEE Proceedings: Control Theory and Applications

SN - 1350-2379

IS - 4

ER -