Satellite attitude control through evolving a neural network

Shuguang Li; Jianping Yuan; Jianjun Luo; Weihua Ma

doi:10.1109/ICMA.2010.5588493

Satellite attitude control through evolving a neural network

Shuguang Li, Jianping Yuan, Jianjun Luo, Weihua Ma

School of Astronautics

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

Abstract

We propose a pure topological recurrent network controller for satellite attitude control, which has random binary connections in hidden layer, and all hidden neurons are activated by sinusoidal functions. A direct graph encoding method and four genetic operators are implemented for using genetic programming to train this controller. Moreover, a simulated small satellite which equipped with three reaction wheels was developed, then this simulator was employed to test the controller and training method for a given simple attitude adjusting mission. The experimental results reveal that this controller has the simplicity, usability and potentials for satellite attitude control through evolutionary learning.

Original language	English
Title of host publication	2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010
Pages	553-559
Number of pages	7
DOIs	https://doi.org/10.1109/ICMA.2010.5588493
State	Published - 2010
Event	2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010 - Xi'an, China Duration: 4 Aug 2010 → 7 Aug 2010

Publication series

Name	2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010

Conference

Conference	2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010
Country/Territory	China
City	Xi'an
Period	4/08/10 → 7/08/10

Access to Document

10.1109/ICMA.2010.5588493

Cite this

@inproceedings{e51c4f3b6f144b14835f7f0c4233e956,

title = "Satellite attitude control through evolving a neural network",

abstract = "We propose a pure topological recurrent network controller for satellite attitude control, which has random binary connections in hidden layer, and all hidden neurons are activated by sinusoidal functions. A direct graph encoding method and four genetic operators are implemented for using genetic programming to train this controller. Moreover, a simulated small satellite which equipped with three reaction wheels was developed, then this simulator was employed to test the controller and training method for a given simple attitude adjusting mission. The experimental results reveal that this controller has the simplicity, usability and potentials for satellite attitude control through evolutionary learning.",

author = "Shuguang Li and Jianping Yuan and Jianjun Luo and Weihua Ma",

year = "2010",

doi = "10.1109/ICMA.2010.5588493",

language = "英语",

isbn = "9781424451418",

series = "2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010",

pages = "553--559",

booktitle = "2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010",

note = "2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010 ; Conference date: 04-08-2010 Through 07-08-2010",

}

Li, S, Yuan, J, Luo, J & Ma, W 2010, Satellite attitude control through evolving a neural network. in 2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010., 5588493, 2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010, pp. 553-559, 2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010, Xi'an, China, 4/08/10. https://doi.org/10.1109/ICMA.2010.5588493

Satellite attitude control through evolving a neural network. / Li, Shuguang; Yuan, Jianping; Luo, Jianjun et al.
2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010. 2010. p. 553-559 5588493 (2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010).

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

TY - GEN

T1 - Satellite attitude control through evolving a neural network

AU - Li, Shuguang

AU - Yuan, Jianping

AU - Luo, Jianjun

AU - Ma, Weihua

PY - 2010

Y1 - 2010

N2 - We propose a pure topological recurrent network controller for satellite attitude control, which has random binary connections in hidden layer, and all hidden neurons are activated by sinusoidal functions. A direct graph encoding method and four genetic operators are implemented for using genetic programming to train this controller. Moreover, a simulated small satellite which equipped with three reaction wheels was developed, then this simulator was employed to test the controller and training method for a given simple attitude adjusting mission. The experimental results reveal that this controller has the simplicity, usability and potentials for satellite attitude control through evolutionary learning.

AB - We propose a pure topological recurrent network controller for satellite attitude control, which has random binary connections in hidden layer, and all hidden neurons are activated by sinusoidal functions. A direct graph encoding method and four genetic operators are implemented for using genetic programming to train this controller. Moreover, a simulated small satellite which equipped with three reaction wheels was developed, then this simulator was employed to test the controller and training method for a given simple attitude adjusting mission. The experimental results reveal that this controller has the simplicity, usability and potentials for satellite attitude control through evolutionary learning.

UR - http://www.scopus.com/inward/record.url?scp=78649259793&partnerID=8YFLogxK

U2 - 10.1109/ICMA.2010.5588493

DO - 10.1109/ICMA.2010.5588493

M3 - 会议稿件

AN - SCOPUS:78649259793

SN - 9781424451418

T3 - 2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010

SP - 553

EP - 559

BT - 2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010

T2 - 2010 IEEE International Conference on Mechatronics and Automation, ICMA 2010

Y2 - 4 August 2010 through 7 August 2010

ER -

Satellite attitude control through evolving a neural network

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this