Prescribed performance fixed-time recurrent neural network control for uncertain nonlinear systems

Junkang Ni, Choon Ki Ahn, Ling Liu, Chongxin Liu

Research output: Contribution to journalArticle

Abstract

This paper investigates fixed-time prescribed performance control problem for uncertain strict-feedback nonlinear systems with unknown dead zone. First, a novel prescribed performance function (PPF) is proposed and a coordinate transformation is employed to transform the prescribed performance constrained system into an unconstrained one. Next, recurrent neural network is introduced to estimate the uncertain dynamics and fixed-time differentiator is utilized to obtain the derivative of virtual control. Then, a fixed-time dynamic surface control is developed to deal with dead zone and guarantee the convergence of the tracking error within a fixed time. Lyapunov stability analysis shows that the presented control scheme can achieve the fixed-time convergence of the error variables, while the other closed-loop system signals are bounded. Finally, numerical simulation validates the effectiveness of the presented control scheme.

Original languageEnglish
JournalNeurocomputing
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Recurrent neural networks
Nonlinear systems
Control surfaces
Closed loop systems
Derivatives
Feedback
Computer simulation

Keywords

  • Dead zone
  • Fixed-time control
  • Prescribed performance control
  • Recurrent neural network control
  • Uncertain nonlinear system

ASJC Scopus subject areas

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Prescribed performance fixed-time recurrent neural network control for uncertain nonlinear systems. / Ni, Junkang; Ahn, Choon Ki; Liu, Ling; Liu, Chongxin.

In: Neurocomputing, 01.01.2019.

Research output: Contribution to journalArticle

@article{0aa4e648177f4ab6b4e820c37693fb40,
title = "Prescribed performance fixed-time recurrent neural network control for uncertain nonlinear systems",
abstract = "This paper investigates fixed-time prescribed performance control problem for uncertain strict-feedback nonlinear systems with unknown dead zone. First, a novel prescribed performance function (PPF) is proposed and a coordinate transformation is employed to transform the prescribed performance constrained system into an unconstrained one. Next, recurrent neural network is introduced to estimate the uncertain dynamics and fixed-time differentiator is utilized to obtain the derivative of virtual control. Then, a fixed-time dynamic surface control is developed to deal with dead zone and guarantee the convergence of the tracking error within a fixed time. Lyapunov stability analysis shows that the presented control scheme can achieve the fixed-time convergence of the error variables, while the other closed-loop system signals are bounded. Finally, numerical simulation validates the effectiveness of the presented control scheme.",
keywords = "Dead zone, Fixed-time control, Prescribed performance control, Recurrent neural network control, Uncertain nonlinear system",
author = "Junkang Ni and Ahn, {Choon Ki} and Ling Liu and Chongxin Liu",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.neucom.2019.07.053",
language = "English",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - Prescribed performance fixed-time recurrent neural network control for uncertain nonlinear systems

AU - Ni, Junkang

AU - Ahn, Choon Ki

AU - Liu, Ling

AU - Liu, Chongxin

PY - 2019/1/1

Y1 - 2019/1/1

N2 - This paper investigates fixed-time prescribed performance control problem for uncertain strict-feedback nonlinear systems with unknown dead zone. First, a novel prescribed performance function (PPF) is proposed and a coordinate transformation is employed to transform the prescribed performance constrained system into an unconstrained one. Next, recurrent neural network is introduced to estimate the uncertain dynamics and fixed-time differentiator is utilized to obtain the derivative of virtual control. Then, a fixed-time dynamic surface control is developed to deal with dead zone and guarantee the convergence of the tracking error within a fixed time. Lyapunov stability analysis shows that the presented control scheme can achieve the fixed-time convergence of the error variables, while the other closed-loop system signals are bounded. Finally, numerical simulation validates the effectiveness of the presented control scheme.

AB - This paper investigates fixed-time prescribed performance control problem for uncertain strict-feedback nonlinear systems with unknown dead zone. First, a novel prescribed performance function (PPF) is proposed and a coordinate transformation is employed to transform the prescribed performance constrained system into an unconstrained one. Next, recurrent neural network is introduced to estimate the uncertain dynamics and fixed-time differentiator is utilized to obtain the derivative of virtual control. Then, a fixed-time dynamic surface control is developed to deal with dead zone and guarantee the convergence of the tracking error within a fixed time. Lyapunov stability analysis shows that the presented control scheme can achieve the fixed-time convergence of the error variables, while the other closed-loop system signals are bounded. Finally, numerical simulation validates the effectiveness of the presented control scheme.

KW - Dead zone

KW - Fixed-time control

KW - Prescribed performance control

KW - Recurrent neural network control

KW - Uncertain nonlinear system

UR - http://www.scopus.com/inward/record.url?scp=85069704341&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069704341&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2019.07.053

DO - 10.1016/j.neucom.2019.07.053

M3 - Article

AN - SCOPUS:85069704341

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -