On using a support vector machine in learning feed-forward control

Bas J. de Kruif; Theo J.A. de Vries

doi:10.1109/AIM.2001.936466

On using a support vector machine in learning feed-forward control

Bas J. de Kruif, Theo J.A. de Vries

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

25 Citations (Scopus)

82 Downloads (Pure)

Abstract

For mechatronic motion systems, the performance increases significantly if, besides feedback control, also feed-forward control is used. This feed-forward part should contain the (stable part of the) inverse of the plant. This inverse is difficult to obtain if non-linear dynamics are present. To overcome this problem, learning feed-forward control can be applied. The properties of the learning mechanism are of importance in this setting. In the paper, a support vector machine is proposed as the learning mechanism. It is shown that this mechanism has several advantages over other learning techniques when applied to learning feed-forward control. The method is tested with simulations

Original language	English
Title of host publication	2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics
Subtitle of host publication	proceedings
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	272-277
Number of pages	6
ISBN (Electronic)	0-7803-6737-5
ISBN (Print)	0-7803-6736-7
DOIs	https://doi.org/10.1109/AIM.2001.936466
Publication status	Published - 1 Jul 2001
Event	2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2001 - Como, Italy Duration: 8 Jul 2001 → 12 Jul 2001

Conference

Conference	2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2001
Abbreviated title	AIM
Country/Territory	Italy
City	Como
Period	8/07/01 → 12/07/01

Access to Document

10.1109/AIM.2001.936466

Cite this

@inproceedings{a9311341ddb7464fb7389ebb312b83fb,

title = "On using a support vector machine in learning feed-forward control",

abstract = "For mechatronic motion systems, the performance increases significantly if, besides feedback control, also feed-forward control is used. This feed-forward part should contain the (stable part of the) inverse of the plant. This inverse is difficult to obtain if non-linear dynamics are present. To overcome this problem, learning feed-forward control can be applied. The properties of the learning mechanism are of importance in this setting. In the paper, a support vector machine is proposed as the learning mechanism. It is shown that this mechanism has several advantages over other learning techniques when applied to learning feed-forward control. The method is tested with simulations",

author = "{de Kruif}, {Bas J.} and {de Vries}, {Theo J.A.}",

year = "2001",

month = jul,

day = "1",

doi = "10.1109/AIM.2001.936466",

language = "English",

isbn = "0-7803-6736-7",

pages = "272--277",

booktitle = "2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics",

publisher = "IEEE",

address = "United States",

note = "2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2001, AIM ; Conference date: 08-07-2001 Through 12-07-2001",

}

de Kruif, BJ & de Vries, TJA 2001, On using a support vector machine in learning feed-forward control. in 2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics: proceedings. IEEE, Piscataway, NJ, pp. 272-277, 2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2001, Como, Italy, 8/07/01. https://doi.org/10.1109/AIM.2001.936466

TY - GEN

T1 - On using a support vector machine in learning feed-forward control

AU - de Kruif, Bas J.

AU - de Vries, Theo J.A.

PY - 2001/7/1

Y1 - 2001/7/1

N2 - For mechatronic motion systems, the performance increases significantly if, besides feedback control, also feed-forward control is used. This feed-forward part should contain the (stable part of the) inverse of the plant. This inverse is difficult to obtain if non-linear dynamics are present. To overcome this problem, learning feed-forward control can be applied. The properties of the learning mechanism are of importance in this setting. In the paper, a support vector machine is proposed as the learning mechanism. It is shown that this mechanism has several advantages over other learning techniques when applied to learning feed-forward control. The method is tested with simulations

AB - For mechatronic motion systems, the performance increases significantly if, besides feedback control, also feed-forward control is used. This feed-forward part should contain the (stable part of the) inverse of the plant. This inverse is difficult to obtain if non-linear dynamics are present. To overcome this problem, learning feed-forward control can be applied. The properties of the learning mechanism are of importance in this setting. In the paper, a support vector machine is proposed as the learning mechanism. It is shown that this mechanism has several advantages over other learning techniques when applied to learning feed-forward control. The method is tested with simulations

U2 - 10.1109/AIM.2001.936466

DO - 10.1109/AIM.2001.936466

M3 - Conference contribution

SN - 0-7803-6736-7

SP - 272

EP - 277

BT - 2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics

PB - IEEE

CY - Piscataway, NJ

T2 - 2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2001

Y2 - 8 July 2001 through 12 July 2001

ER -

On using a support vector machine in learning feed-forward control

Abstract

Conference

Access to Document

Fingerprint

Cite this