Output details
13 - Electrical and Electronic Engineering, Metallurgy and Materials
University of Southampton
Article title
NARX-based nonlinear system identification using orthogonal least squares basis hunting
Type
D - Journal article
Title of journal
IEEE Transactions on Control Systems Technology
Article number
-
Volume number
16
Issue number
1
First page of article
78
ISSN of journal
1063-6536
Year of publication
2008
Number of additional authors
2
Additional information
Significance of output:
A significant enhancement to the celebrated orthogonal least squares algorithm for radial basis function modelling was developed, in which a computational intelligence algorithm is used to select model structure. The advantages of this approach are better generalisation capability, a more parsimonious model and no need for tuning learning hyperparameters. The wide applicability of the approach has resulted in it being adopted in spheres ranging from ship control to polymer rheology.
Interdisciplinary
-
Cross-referral requested
-
Research group
None
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
