A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation

Qiao Lin; Qinglai Wei; Bo Zhao

doi:10.1007/978-3-319-59081-3_8

A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation

Qiao Lin, Qinglai Wei, Bo Zhao

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

In this study, a nonquadratic performance function is introduced to overcome the saturation nonlinearity in actuators. Then a novel solution, generalized policy iteration adaptive dynamic programming algorithm, is applied to deal with the problem of optimal control. To achieve this goal, we use two neural networks to approximate control vectors and performance index function. Finally, this paper focuses on an example simulated on Matlab, which verifies the excellent convergence of the mentioned algorithm and feasibility of this scheme.

Original language	English
Title of host publication	Advances in Neural Networks - ISNN 2017 - 14th International Symposium, ISNN 2017, Proceedings
Editors	Fengyu Cong, Qinglai Wei, Andrew Leung
Publisher	Springer Verlag
Pages	60-65
Number of pages	6
ISBN (Print)	9783319590806
DOIs	https://doi.org/10.1007/978-3-319-59081-3_8
Publication status	Published - 2017
Externally published	Yes
Event	14th International Symposium on Neural Networks, ISNN 2017 - Sapporo, Hakodate, and Muroran, Hokkaido, Japan Duration: 21 Jun 2017 → 26 Jun 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10262 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	14th International Symposium on Neural Networks, ISNN 2017
Country/Territory	Japan
City	Sapporo, Hakodate, and Muroran, Hokkaido
Period	21/06/17 → 26/06/17

Keywords

Adaptive dynamic programming
Neural network
Optimal control
Saturating actuators

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-319-59081-3_8

http://link.springer.com/10.1007/978-3-319-59081-3_8

Cite this

Lin, Q., Wei, Q., & Zhao, B. (2017). A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation. In F. Cong, Q. Wei, & A. Leung (Eds.), Advances in Neural Networks - ISNN 2017 - 14th International Symposium, ISNN 2017, Proceedings (pp. 60-65). Article Chapter 8 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10262 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-59081-3_8

Lin, Qiao ; Wei, Qinglai ; Zhao, Bo. / A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation. Advances in Neural Networks - ISNN 2017 - 14th International Symposium, ISNN 2017, Proceedings. editor / Fengyu Cong ; Qinglai Wei ; Andrew Leung. Springer Verlag, 2017. pp. 60-65 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{3860829e55c443be845029e0c5e22cf0,

title = "A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation",

abstract = "In this study, a nonquadratic performance function is introduced to overcome the saturation nonlinearity in actuators. Then a novel solution, generalized policy iteration adaptive dynamic programming algorithm, is applied to deal with the problem of optimal control. To achieve this goal, we use two neural networks to approximate control vectors and performance index function. Finally, this paper focuses on an example simulated on Matlab, which verifies the excellent convergence of the mentioned algorithm and feasibility of this scheme.",

keywords = "Adaptive dynamic programming, Neural network, Optimal control, Saturating actuators",

author = "Qiao Lin and Qinglai Wei and Bo Zhao",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 14th International Symposium on Neural Networks, ISNN 2017 ; Conference date: 21-06-2017 Through 26-06-2017",

year = "2017",

doi = "10.1007/978-3-319-59081-3_8",

language = "English",

isbn = "9783319590806",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "60--65",

editor = "Fengyu Cong and Qinglai Wei and Andrew Leung",

booktitle = "Advances in Neural Networks - ISNN 2017 - 14th International Symposium, ISNN 2017, Proceedings",

address = "Germany",

}

Lin, Q, Wei, Q & Zhao, B 2017, A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation. in F Cong, Q Wei & A Leung (eds), Advances in Neural Networks - ISNN 2017 - 14th International Symposium, ISNN 2017, Proceedings., Chapter 8, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10262 LNCS, Springer Verlag, pp. 60-65, 14th International Symposium on Neural Networks, ISNN 2017, Sapporo, Hakodate, and Muroran, Hokkaido, Japan, 21/06/17. https://doi.org/10.1007/978-3-319-59081-3_8

A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation. / Lin, Qiao; Wei, Qinglai; Zhao, Bo.
Advances in Neural Networks - ISNN 2017 - 14th International Symposium, ISNN 2017, Proceedings. ed. / Fengyu Cong; Qinglai Wei; Andrew Leung. Springer Verlag, 2017. p. 60-65 Chapter 8 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10262 LNCS).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation

AU - Lin, Qiao

AU - Wei, Qinglai

AU - Zhao, Bo

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - In this study, a nonquadratic performance function is introduced to overcome the saturation nonlinearity in actuators. Then a novel solution, generalized policy iteration adaptive dynamic programming algorithm, is applied to deal with the problem of optimal control. To achieve this goal, we use two neural networks to approximate control vectors and performance index function. Finally, this paper focuses on an example simulated on Matlab, which verifies the excellent convergence of the mentioned algorithm and feasibility of this scheme.

AB - In this study, a nonquadratic performance function is introduced to overcome the saturation nonlinearity in actuators. Then a novel solution, generalized policy iteration adaptive dynamic programming algorithm, is applied to deal with the problem of optimal control. To achieve this goal, we use two neural networks to approximate control vectors and performance index function. Finally, this paper focuses on an example simulated on Matlab, which verifies the excellent convergence of the mentioned algorithm and feasibility of this scheme.

KW - Adaptive dynamic programming

KW - Neural network

KW - Optimal control

KW - Saturating actuators

UR - http://www.scopus.com/inward/record.url?scp=85021743576&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-59081-3_8

DO - 10.1007/978-3-319-59081-3_8

M3 - Conference contribution

SN - 9783319590806

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 60

EP - 65

BT - Advances in Neural Networks - ISNN 2017 - 14th International Symposium, ISNN 2017, Proceedings

A2 - Cong, Fengyu

A2 - Wei, Qinglai

A2 - Leung, Andrew

PB - Springer Verlag

T2 - 14th International Symposium on Neural Networks, ISNN 2017

Y2 - 21 June 2017 through 26 June 2017

ER -

Lin Q, Wei Q, Zhao B. A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation. In Cong F, Wei Q, Leung A, editors, Advances in Neural Networks - ISNN 2017 - 14th International Symposium, ISNN 2017, Proceedings. Springer Verlag. 2017. p. 60-65. Chapter 8. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-59081-3_8

A generalized policy iteration adaptive dynamic programming algorithm for optimal control of discrete-time nonlinear systems with actuator saturation

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this