Selection of optimal learning rates in CMAC based control schemes

Wen Chi Luo; Kai-Tai Song

doi:10.1109/ISIC.2002.1157764

Selection of optimal learning rates in CMAC based control schemes

Wen Chi Luo^*, Kai-Tai Song

^*Corresponding author for this work

Department of Electrical and Computer Engineering

Research output: Contribution to conference › Paper › peer-review

1 Scopus citations

Abstract

CMAC based control schemes have been studied by many researchers. It is well recognized that properly designed CMAC controllers provide useful and practical tools for precision control of non-linear systems. For complex trajectories, however, the convergence speed of CMAC can be slow because the CMAC module takes much time in learning the inverse dynamics of the plant. Therefore, one practical difficulty of CMAC based controller design is the selection of appropriate learning rate. In this paper, we present a method for selection of optimal CMAC learning rate. Furthermore, we demonstrate that the proposed GA-based approach to parameter selection can provide a global optimal solution. Computer simulation results confirm the effectiveness of the proposed method.

Original language	English
Pages	212-216
Number of pages	5
DOIs	https://doi.org/10.1109/ISIC.2002.1157764
State	Published - 30 Oct 2002
Event	Proceedings of the 2002 IEEE International Symposium on Intelligent Control - Vancouver, Canada Duration: 27 Oct 2002 → 30 Oct 2002

Conference

Conference	Proceedings of the 2002 IEEE International Symposium on Intelligent Control
Country/Territory	Canada
City	Vancouver
Period	27/10/02 → 30/10/02

Keywords

Artificial neural networks
Genetic algorithms
Learning control
Parameter optimization

Access to Document

10.1109/ISIC.2002.1157764

Cite this

@conference{294f72e977314cc398868575af769f63,

title = "Selection of optimal learning rates in CMAC based control schemes",

abstract = "CMAC based control schemes have been studied by many researchers. It is well recognized that properly designed CMAC controllers provide useful and practical tools for precision control of non-linear systems. For complex trajectories, however, the convergence speed of CMAC can be slow because the CMAC module takes much time in learning the inverse dynamics of the plant. Therefore, one practical difficulty of CMAC based controller design is the selection of appropriate learning rate. In this paper, we present a method for selection of optimal CMAC learning rate. Furthermore, we demonstrate that the proposed GA-based approach to parameter selection can provide a global optimal solution. Computer simulation results confirm the effectiveness of the proposed method.",

keywords = "Artificial neural networks, Genetic algorithms, Learning control, Parameter optimization",

author = "Luo, {Wen Chi} and Kai-Tai Song",

year = "2002",

month = oct,

day = "30",

doi = "10.1109/ISIC.2002.1157764",

language = "English",

pages = "212--216",

note = "Proceedings of the 2002 IEEE International Symposium on Intelligent Control ; Conference date: 27-10-2002 Through 30-10-2002",

}

TY - CONF

T1 - Selection of optimal learning rates in CMAC based control schemes

AU - Luo, Wen Chi

AU - Song, Kai-Tai

PY - 2002/10/30

Y1 - 2002/10/30

N2 - CMAC based control schemes have been studied by many researchers. It is well recognized that properly designed CMAC controllers provide useful and practical tools for precision control of non-linear systems. For complex trajectories, however, the convergence speed of CMAC can be slow because the CMAC module takes much time in learning the inverse dynamics of the plant. Therefore, one practical difficulty of CMAC based controller design is the selection of appropriate learning rate. In this paper, we present a method for selection of optimal CMAC learning rate. Furthermore, we demonstrate that the proposed GA-based approach to parameter selection can provide a global optimal solution. Computer simulation results confirm the effectiveness of the proposed method.

AB - CMAC based control schemes have been studied by many researchers. It is well recognized that properly designed CMAC controllers provide useful and practical tools for precision control of non-linear systems. For complex trajectories, however, the convergence speed of CMAC can be slow because the CMAC module takes much time in learning the inverse dynamics of the plant. Therefore, one practical difficulty of CMAC based controller design is the selection of appropriate learning rate. In this paper, we present a method for selection of optimal CMAC learning rate. Furthermore, we demonstrate that the proposed GA-based approach to parameter selection can provide a global optimal solution. Computer simulation results confirm the effectiveness of the proposed method.

KW - Artificial neural networks

KW - Genetic algorithms

KW - Learning control

KW - Parameter optimization

UR - http://www.scopus.com/inward/record.url?scp=0036916557&partnerID=8YFLogxK

U2 - 10.1109/ISIC.2002.1157764

DO - 10.1109/ISIC.2002.1157764

M3 - Paper

AN - SCOPUS:0036916557

SP - 212

EP - 216

T2 - Proceedings of the 2002 IEEE International Symposium on Intelligent Control

Y2 - 27 October 2002 through 30 October 2002

ER -

Selection of optimal learning rates in CMAC based control schemes

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this