On-Line Modeling of AUV’s Maneuvering Motion Based on Sliding Window SVM

DOI

10.2991/icaise.2013.2How to use a DOI?

Keywords

least squares support vector machines, sliding window, parametric identification, free Running model test

Abstract

In order to solve the problem of on-line modeling of AUV's Maneuvering Motion, considering that the sample space expends in the parametric identification, a learning algorithm for least squares support vector machines (LS-SVM) based on sliding window is deduced. The method improves the identification results of SVM and has the merits such as computing fast and cost less storage space. Firstly, based on the theory of sliding window, a data region which is sliding with time is built to restrain matrix dimension expending in the parametric identification. Then, the parameters of a certain AUV are identified on-line based on the free running model test. And the veracity of the parameters identified is validated by using the system simulation. The results demonstrate that the method which can identify the parameters effectively is feasible and effective.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yongjie Pang
AU  - Guocheng Zhang
AU  - Zhonghui Hu
AU  - Yushan Sun
PY  - 2013/08
DA  - 2013/08
TI  - On-Line Modeling of AUV’s Maneuvering Motion Based on Sliding Window SVM
BT  - Proceedings of the 2013 The International Conference on Artificial Intelligence and Software Engineering (ICAISE 2013)
PB  - Atlantis Press
SP  - 5
EP  - 10
SN  - 1951-6851
UR  - https://doi.org/10.2991/icaise.2013.2
DO  - 10.2991/icaise.2013.2
ID  - Pang2013/08
ER  -