Recurrent wavelet-based Elman neural network control for multi-axis motion control stage using linear ultrasonic motors

F. J. Lin, Y. S. Kung, Syuan-Yi Chen, Y. H. Liu

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

A novel recurrent wavelet-based Elman neural network (RWENN) control system is proposed in this study to control the mover position of a multi-axis motion control stage using linear ultrasonic motors (LUSMs) for the tracking of various contours. First, the structure and operating principles of the LUSMs are introduced briefly. Since the dynamic characteristics and motor parameters of the LUSMs are non-linear and time varying, the RWENN is proposed to control the mover of the X-Y-θ motion control stage to track various contours precisely using a direct decentralised control strategy. In the proposed RWENN, each hidden neuron employs a different wavelet function as an activation function. Moreover, the recurrent connective weights are added in the RWENN. Therefore compared with the conventional Elman neural network (ENN), both the precision and time of convergence are improved. Furthermore, the on-line learning algorithm based on the supervised gradient descent method and the convergence analysis of the tracking error using a discrete-type Lyapunov function of the RWENN are developed. Finally, some experimental results of various contours tracking show that the tracking performance of the RWENN is significantly improved compared with the ENN.

Original languageEnglish
Article numberIEPAAN000004000005000314000001
Pages (from-to)314-332
Number of pages19
JournalIET Electric Power Applications
Volume4
Issue number5
DOIs
Publication statusPublished - 2010 May 1

Fingerprint

Motion control
Ultrasonics
Neural networks
Decentralized control
Lyapunov functions
Learning algorithms
Neurons
Chemical activation
Control systems

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Recurrent wavelet-based Elman neural network control for multi-axis motion control stage using linear ultrasonic motors. / Lin, F. J.; Kung, Y. S.; Chen, Syuan-Yi; Liu, Y. H.

In: IET Electric Power Applications, Vol. 4, No. 5, IEPAAN000004000005000314000001, 01.05.2010, p. 314-332.

Research output: Contribution to journalArticle

@article{0103af10dc8042d480218f75c7e3e458,
title = "Recurrent wavelet-based Elman neural network control for multi-axis motion control stage using linear ultrasonic motors",
abstract = "A novel recurrent wavelet-based Elman neural network (RWENN) control system is proposed in this study to control the mover position of a multi-axis motion control stage using linear ultrasonic motors (LUSMs) for the tracking of various contours. First, the structure and operating principles of the LUSMs are introduced briefly. Since the dynamic characteristics and motor parameters of the LUSMs are non-linear and time varying, the RWENN is proposed to control the mover of the X-Y-θ motion control stage to track various contours precisely using a direct decentralised control strategy. In the proposed RWENN, each hidden neuron employs a different wavelet function as an activation function. Moreover, the recurrent connective weights are added in the RWENN. Therefore compared with the conventional Elman neural network (ENN), both the precision and time of convergence are improved. Furthermore, the on-line learning algorithm based on the supervised gradient descent method and the convergence analysis of the tracking error using a discrete-type Lyapunov function of the RWENN are developed. Finally, some experimental results of various contours tracking show that the tracking performance of the RWENN is significantly improved compared with the ENN.",
author = "Lin, {F. J.} and Kung, {Y. S.} and Syuan-Yi Chen and Liu, {Y. H.}",
year = "2010",
month = "5",
day = "1",
doi = "10.1049/iet-epa.2009.0144",
language = "English",
volume = "4",
pages = "314--332",
journal = "IET Electric Power Applications",
issn = "1751-8660",
publisher = "Institution of Engineering and Technology",
number = "5",

}

TY - JOUR

T1 - Recurrent wavelet-based Elman neural network control for multi-axis motion control stage using linear ultrasonic motors

AU - Lin, F. J.

AU - Kung, Y. S.

AU - Chen, Syuan-Yi

AU - Liu, Y. H.

PY - 2010/5/1

Y1 - 2010/5/1

N2 - A novel recurrent wavelet-based Elman neural network (RWENN) control system is proposed in this study to control the mover position of a multi-axis motion control stage using linear ultrasonic motors (LUSMs) for the tracking of various contours. First, the structure and operating principles of the LUSMs are introduced briefly. Since the dynamic characteristics and motor parameters of the LUSMs are non-linear and time varying, the RWENN is proposed to control the mover of the X-Y-θ motion control stage to track various contours precisely using a direct decentralised control strategy. In the proposed RWENN, each hidden neuron employs a different wavelet function as an activation function. Moreover, the recurrent connective weights are added in the RWENN. Therefore compared with the conventional Elman neural network (ENN), both the precision and time of convergence are improved. Furthermore, the on-line learning algorithm based on the supervised gradient descent method and the convergence analysis of the tracking error using a discrete-type Lyapunov function of the RWENN are developed. Finally, some experimental results of various contours tracking show that the tracking performance of the RWENN is significantly improved compared with the ENN.

AB - A novel recurrent wavelet-based Elman neural network (RWENN) control system is proposed in this study to control the mover position of a multi-axis motion control stage using linear ultrasonic motors (LUSMs) for the tracking of various contours. First, the structure and operating principles of the LUSMs are introduced briefly. Since the dynamic characteristics and motor parameters of the LUSMs are non-linear and time varying, the RWENN is proposed to control the mover of the X-Y-θ motion control stage to track various contours precisely using a direct decentralised control strategy. In the proposed RWENN, each hidden neuron employs a different wavelet function as an activation function. Moreover, the recurrent connective weights are added in the RWENN. Therefore compared with the conventional Elman neural network (ENN), both the precision and time of convergence are improved. Furthermore, the on-line learning algorithm based on the supervised gradient descent method and the convergence analysis of the tracking error using a discrete-type Lyapunov function of the RWENN are developed. Finally, some experimental results of various contours tracking show that the tracking performance of the RWENN is significantly improved compared with the ENN.

UR - http://www.scopus.com/inward/record.url?scp=77953514800&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953514800&partnerID=8YFLogxK

U2 - 10.1049/iet-epa.2009.0144

DO - 10.1049/iet-epa.2009.0144

M3 - Article

AN - SCOPUS:77953514800

VL - 4

SP - 314

EP - 332

JO - IET Electric Power Applications

JF - IET Electric Power Applications

SN - 1751-8660

IS - 5

M1 - IEPAAN000004000005000314000001

ER -