NARX network for dynamic system identification in adaptive inverse control

Ya Qiu Liu; Guang Fu Ma; Zhong Shi

NARX network for dynamic system identification in adaptive inverse control

Ya Qiu Liu^*
, Guang Fu Ma
, Zhong Shi

^*Corresponding author for this work

School of Astronautics

Harbin Institute of Technology
Jiamusi University

Research output: Contribution to journal › Article › peer-review

8 Scopus citations

Abstract

According to the learning algorithm of dynamic neural network, an improved real times real life (RTRL algorithm) is presented for Nonlinear Auto-Regressive eXogenous input (NARX) network. Based on LM algorithm, the improved algorithm substitutes for the traditional gradient algorithm and is also applied to dynamic system identification in adaptive inverse control. Practical simulation results show that the NARX dynamic neural network is of great capability to depict the dynamic system, and the given algorithm can improve the convergence speed of learning and is feasible and effective.

Original language	English
Pages (from-to)	173-176
Number of pages	4
Journal	Harbin Gongye Daxue Xuebao/Journal of Harbin Institute of Technology
Volume	37
Issue number	2
State	Published - Feb 2005

Keywords

Adaptive inverse control
Dynamic neural network
NARX network
RTRL algorithm

Cite this

@article{0072de1c403b4bb6b7aaa08a07ba5546,

title = "NARX network for dynamic system identification in adaptive inverse control",

abstract = "According to the learning algorithm of dynamic neural network, an improved real times real life (RTRL algorithm) is presented for Nonlinear Auto-Regressive eXogenous input (NARX) network. Based on LM algorithm, the improved algorithm substitutes for the traditional gradient algorithm and is also applied to dynamic system identification in adaptive inverse control. Practical simulation results show that the NARX dynamic neural network is of great capability to depict the dynamic system, and the given algorithm can improve the convergence speed of learning and is feasible and effective.",

keywords = "Adaptive inverse control, Dynamic neural network, NARX network, RTRL algorithm",

author = "Liu, \{Ya Qiu\} and Ma, \{Guang Fu\} and Zhong Shi",

year = "2005",

month = feb,

language = "英语",

volume = "37",

pages = "173--176",

journal = "Harbin Gongye Daxue Xuebao/Journal of Harbin Institute of Technology",

issn = "0367-6234",

publisher = "Harbin Institute of Technology",

number = "2",

}

TY - JOUR

T1 - NARX network for dynamic system identification in adaptive inverse control

AU - Liu, Ya Qiu

AU - Ma, Guang Fu

AU - Shi, Zhong

PY - 2005/2

Y1 - 2005/2

N2 - According to the learning algorithm of dynamic neural network, an improved real times real life (RTRL algorithm) is presented for Nonlinear Auto-Regressive eXogenous input (NARX) network. Based on LM algorithm, the improved algorithm substitutes for the traditional gradient algorithm and is also applied to dynamic system identification in adaptive inverse control. Practical simulation results show that the NARX dynamic neural network is of great capability to depict the dynamic system, and the given algorithm can improve the convergence speed of learning and is feasible and effective.

AB - According to the learning algorithm of dynamic neural network, an improved real times real life (RTRL algorithm) is presented for Nonlinear Auto-Regressive eXogenous input (NARX) network. Based on LM algorithm, the improved algorithm substitutes for the traditional gradient algorithm and is also applied to dynamic system identification in adaptive inverse control. Practical simulation results show that the NARX dynamic neural network is of great capability to depict the dynamic system, and the given algorithm can improve the convergence speed of learning and is feasible and effective.

KW - Adaptive inverse control

KW - Dynamic neural network

KW - NARX network

KW - RTRL algorithm

UR - https://www.scopus.com/pages/publications/17244365348

M3 - 文章

AN - SCOPUS:17244365348

SN - 0367-6234

VL - 37

SP - 173

EP - 176

JO - Harbin Gongye Daxue Xuebao/Journal of Harbin Institute of Technology

JF - Harbin Gongye Daxue Xuebao/Journal of Harbin Institute of Technology

IS - 2

ER -

NARX network for dynamic system identification in adaptive inverse control

Abstract

Keywords

Other files and links

Fingerprint

Cite this