Investigation of steering dynamics ship model identification based on PSO-LSSVR

Sheng Liu; Jia Song; Bing Li; Gaoyun Li

doi:10.1109/ISSCAA.2008.4776236

Investigation of steering dynamics ship model identification based on PSO-LSSVR

Sheng Liu^*
, Jia Song
, Bing Li
, Gaoyun Li

^*Corresponding author for this work

Harbin Engineering University

Research output: Contribution to conference › Paper › peer-review

5 Scopus citations

Abstract

According to the high-order nonlinearity and parameter uncertainty of the ship steering dynamics, it is difficult to establish the accurate mathematical model by using normal identification methods. To solve this problem, a new kind of Least Squares Support Vector Regression based on the Particle Swarm Optimization (PSO-LSSVR) is proposed. This method can select the parameters of LSSVR automatically without trial and error, thus ensure the accuracy of parameters optimization. Apply this method to the model identification of the ship steering dynamics, and compare the identification effect with the experimental reference data. The PSO-LSSVR is able to establish the system model effectively, the structure is simple and generalization ability is well.

Original language	English
DOIs	https://doi.org/10.1109/ISSCAA.2008.4776236
State	Published - 2008
Externally published	Yes
Event	2008 2nd International Symposium on Systems and Control in Aerospace and Astronautics, ISSCAA 2008 - Shenzhen, China Duration: 10 Dec 2008 → 12 Dec 2008

Conference

Conference	2008 2nd International Symposium on Systems and Control in Aerospace and Astronautics, ISSCAA 2008
Country/Territory	China
City	Shenzhen
Period	10/12/08 → 12/12/08

Keywords

Least squares support vector regression
Nonlinear system identification
Particle swarm optimization
Ship maneuvering

Access to Document

10.1109/ISSCAA.2008.4776236

Cite this

@conference{82caf896090f428d88626e0bef797b51,

title = "Investigation of steering dynamics ship model identification based on PSO-LSSVR",

abstract = "According to the high-order nonlinearity and parameter uncertainty of the ship steering dynamics, it is difficult to establish the accurate mathematical model by using normal identification methods. To solve this problem, a new kind of Least Squares Support Vector Regression based on the Particle Swarm Optimization (PSO-LSSVR) is proposed. This method can select the parameters of LSSVR automatically without trial and error, thus ensure the accuracy of parameters optimization. Apply this method to the model identification of the ship steering dynamics, and compare the identification effect with the experimental reference data. The PSO-LSSVR is able to establish the system model effectively, the structure is simple and generalization ability is well.",

keywords = "Least squares support vector regression, Nonlinear system identification, Particle swarm optimization, Ship maneuvering",

author = "Sheng Liu and Jia Song and Bing Li and Gaoyun Li",

year = "2008",

doi = "10.1109/ISSCAA.2008.4776236",

language = "英语",

note = "2008 2nd International Symposium on Systems and Control in Aerospace and Astronautics, ISSCAA 2008 ; Conference date: 10-12-2008 Through 12-12-2008",

}

TY - CONF

T1 - Investigation of steering dynamics ship model identification based on PSO-LSSVR

AU - Liu, Sheng

AU - Song, Jia

AU - Li, Bing

AU - Li, Gaoyun

PY - 2008

Y1 - 2008

N2 - According to the high-order nonlinearity and parameter uncertainty of the ship steering dynamics, it is difficult to establish the accurate mathematical model by using normal identification methods. To solve this problem, a new kind of Least Squares Support Vector Regression based on the Particle Swarm Optimization (PSO-LSSVR) is proposed. This method can select the parameters of LSSVR automatically without trial and error, thus ensure the accuracy of parameters optimization. Apply this method to the model identification of the ship steering dynamics, and compare the identification effect with the experimental reference data. The PSO-LSSVR is able to establish the system model effectively, the structure is simple and generalization ability is well.

AB - According to the high-order nonlinearity and parameter uncertainty of the ship steering dynamics, it is difficult to establish the accurate mathematical model by using normal identification methods. To solve this problem, a new kind of Least Squares Support Vector Regression based on the Particle Swarm Optimization (PSO-LSSVR) is proposed. This method can select the parameters of LSSVR automatically without trial and error, thus ensure the accuracy of parameters optimization. Apply this method to the model identification of the ship steering dynamics, and compare the identification effect with the experimental reference data. The PSO-LSSVR is able to establish the system model effectively, the structure is simple and generalization ability is well.

KW - Least squares support vector regression

KW - Nonlinear system identification

KW - Particle swarm optimization

KW - Ship maneuvering

UR - https://www.scopus.com/pages/publications/66449090368

U2 - 10.1109/ISSCAA.2008.4776236

DO - 10.1109/ISSCAA.2008.4776236

M3 - 论文

AN - SCOPUS:66449090368

T2 - 2008 2nd International Symposium on Systems and Control in Aerospace and Astronautics, ISSCAA 2008

Y2 - 10 December 2008 through 12 December 2008

ER -

Investigation of steering dynamics ship model identification based on PSO-LSSVR

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this