Share Email Print

Proceedings Paper

Accelerated convergence of neural network system identification algorithms via principal component analysis
Author(s): David C. Hyland; Lawrence D. Davis; Keith K. Denoyer
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

While significant theoretical and experimental progress has been made in the development of neural network-based systems for the autonomous identification and control of space platforms, there remain important unresolved issues associated with the reliable prediction of convergence speed and the avoidance of inordinately slow convergence. To speed convergence of neural identifiers, we introduce the preprocessing of identifier inputs using Principal Component Analysis (PCA) algorithms. Which automatically transform the neural identifier's external inputs so as to make the correlation matrix identity, resulting in enormous improvements in the convergence speed of the neural identifier. From a study of several such algorithms, we developed a new PCA approach which exhibits excellent convergence properties, insensitivity to noise and reliable accuracy.

Paper Details

Date Published: 8 December 1998
PDF: 12 pages
Proc. SPIE 3430, Novel Optical Systems and Large-Aperture Imaging, (8 December 1998); doi: 10.1117/12.332480
Show Author Affiliations
David C. Hyland, Univ. of Michigan (United States)
Lawrence D. Davis, Planning Systems, Inc. (United States)
Keith K. Denoyer, Air Force Research Lab. (United States)

Published in SPIE Proceedings Vol. 3430:
Novel Optical Systems and Large-Aperture Imaging
Kevin Dean Bell; Michael K. Powers; Jose M. Sasian, Editor(s)

© SPIE. Terms of Use
Back to Top