A new noise-tolerant and predefined-time ZNN model for time-dependent matrix inversion

Lin Xiao, Yongsheng Zhang, Jianhua Dai, Ke Chen, Song Yang, Weibing Li, Bolin Liao, Lei Ding, Jichun Li

Research output: Contribution to journalArticleResearchpeer-review

Abstract

In this work, a new zeroing neural network (ZNN) using a versatile activation function (VAF) is presented and introduced for solving time-dependent matrix inversion. Unlike existing ZNN models, the proposed ZNN model not only converges to zero within a predefined finite time but also tolerates several noises in solving the time-dependent matrix inversion, and thus called new noise-tolerant ZNN (NNTZNN) model. In addition, the convergence and robustness of this model are mathematically analyzed in detail. Two comparative numerical simulations with different dimensions are used to test the efficiency and superiority of the NNTZNN model to the previous ZNN models using other activation functions. In addition, two practical application examples (i.e., a mobile manipulator and a real Kinova JACO2 robot manipulator) are presented to validate the applicability and physical feasibility of the NNTZNN model in a noisy environment. Both simulative and experimental results demonstrate the effectiveness and tolerant-noise ability of the NNTZNN model.
Original languageEnglish
Pages (from-to)124-134
Number of pages11
JournalNeural Networks
Volume117
Early online date15 May 2019
DOIs
Publication statusPublished - 30 Sep 2019

Fingerprint

Neural Networks (Computer)
Noise
Neural networks
Manipulators
Chemical activation
Robots
Computer simulation

Cite this

Xiao, Lin ; Zhang, Yongsheng ; Dai, Jianhua ; Chen, Ke ; Yang, Song ; Li, Weibing ; Liao, Bolin ; Ding, Lei ; Li, Jichun. / A new noise-tolerant and predefined-time ZNN model for time-dependent matrix inversion. In: Neural Networks. 2019 ; Vol. 117. pp. 124-134.
@article{d440700ed31c49d6ac24ca02af6df9d1,
title = "A new noise-tolerant and predefined-time ZNN model for time-dependent matrix inversion",
abstract = "In this work, a new zeroing neural network (ZNN) using a versatile activation function (VAF) is presented and introduced for solving time-dependent matrix inversion. Unlike existing ZNN models, the proposed ZNN model not only converges to zero within a predefined finite time but also tolerates several noises in solving the time-dependent matrix inversion, and thus called new noise-tolerant ZNN (NNTZNN) model. In addition, the convergence and robustness of this model are mathematically analyzed in detail. Two comparative numerical simulations with different dimensions are used to test the efficiency and superiority of the NNTZNN model to the previous ZNN models using other activation functions. In addition, two practical application examples (i.e., a mobile manipulator and a real Kinova JACO2 robot manipulator) are presented to validate the applicability and physical feasibility of the NNTZNN model in a noisy environment. Both simulative and experimental results demonstrate the effectiveness and tolerant-noise ability of the NNTZNN model.",
author = "Lin Xiao and Yongsheng Zhang and Jianhua Dai and Ke Chen and Song Yang and Weibing Li and Bolin Liao and Lei Ding and Jichun Li",
year = "2019",
month = "9",
day = "30",
doi = "10.1016/j.neunet.2019.05.005",
language = "English",
volume = "117",
pages = "124--134",
journal = "Neural Networks",
issn = "0893-6080",
publisher = "Elsevier",

}

Xiao, L, Zhang, Y, Dai, J, Chen, K, Yang, S, Li, W, Liao, B, Ding, L & Li, J 2019, 'A new noise-tolerant and predefined-time ZNN model for time-dependent matrix inversion', Neural Networks, vol. 117, pp. 124-134. https://doi.org/10.1016/j.neunet.2019.05.005

A new noise-tolerant and predefined-time ZNN model for time-dependent matrix inversion. / Xiao, Lin; Zhang, Yongsheng; Dai, Jianhua; Chen, Ke; Yang, Song; Li, Weibing; Liao, Bolin; Ding, Lei; Li, Jichun.

In: Neural Networks, Vol. 117, 30.09.2019, p. 124-134.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - A new noise-tolerant and predefined-time ZNN model for time-dependent matrix inversion

AU - Xiao, Lin

AU - Zhang, Yongsheng

AU - Dai, Jianhua

AU - Chen, Ke

AU - Yang, Song

AU - Li, Weibing

AU - Liao, Bolin

AU - Ding, Lei

AU - Li, Jichun

PY - 2019/9/30

Y1 - 2019/9/30

N2 - In this work, a new zeroing neural network (ZNN) using a versatile activation function (VAF) is presented and introduced for solving time-dependent matrix inversion. Unlike existing ZNN models, the proposed ZNN model not only converges to zero within a predefined finite time but also tolerates several noises in solving the time-dependent matrix inversion, and thus called new noise-tolerant ZNN (NNTZNN) model. In addition, the convergence and robustness of this model are mathematically analyzed in detail. Two comparative numerical simulations with different dimensions are used to test the efficiency and superiority of the NNTZNN model to the previous ZNN models using other activation functions. In addition, two practical application examples (i.e., a mobile manipulator and a real Kinova JACO2 robot manipulator) are presented to validate the applicability and physical feasibility of the NNTZNN model in a noisy environment. Both simulative and experimental results demonstrate the effectiveness and tolerant-noise ability of the NNTZNN model.

AB - In this work, a new zeroing neural network (ZNN) using a versatile activation function (VAF) is presented and introduced for solving time-dependent matrix inversion. Unlike existing ZNN models, the proposed ZNN model not only converges to zero within a predefined finite time but also tolerates several noises in solving the time-dependent matrix inversion, and thus called new noise-tolerant ZNN (NNTZNN) model. In addition, the convergence and robustness of this model are mathematically analyzed in detail. Two comparative numerical simulations with different dimensions are used to test the efficiency and superiority of the NNTZNN model to the previous ZNN models using other activation functions. In addition, two practical application examples (i.e., a mobile manipulator and a real Kinova JACO2 robot manipulator) are presented to validate the applicability and physical feasibility of the NNTZNN model in a noisy environment. Both simulative and experimental results demonstrate the effectiveness and tolerant-noise ability of the NNTZNN model.

UR - http://www.scopus.com/inward/record.url?scp=85066332149&partnerID=8YFLogxK

U2 - 10.1016/j.neunet.2019.05.005

DO - 10.1016/j.neunet.2019.05.005

M3 - Article

VL - 117

SP - 124

EP - 134

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

ER -