DocumentCode :
2865765
Title :
High Order B-Spline Networks and Its Applications to Learning Feedforward Control
Author :
Song, Zhen ; Chen, YangQuan
Author_Institution :
Dept. of Electr. & Comput. Eng., Utah State Univ., Logan, UT
fYear :
2006
fDate :
25-28 June 2006
Firstpage :
666
Lastpage :
671
Abstract :
This paper presents a derivation of closed-form expression of an arbitrary high order uniform B-spline. This close-form expression is then applied to the stability analysis of a learning feedforward controller (LFFC) using a B-spline network (BSN) in frequency domain. BSN as a promising tool discussed in the existing literature is mainly limited to the 2nd-order B-splines case. Detailed derivation of high order BSN is still missing from the literature. In this paper, we propose to use a special type of matrix called "B-matrix" to facilitate the analysis and computation of high-order BSN for use with the LFFC scheme. Unlike the common recursive definition, the technique we developed is in "piecewise closed form," which is a nice feature especially suitable for the LFFC analysis and design
Keywords :
adaptive control; control system synthesis; curve fitting; feedforward; iterative methods; learning systems; piecewise polynomial techniques; splines (mathematics); stability; B-matrix; frequency domain; high order B-spline networks; learning feedforward control; piecewise closed form; stability analysis; Artificial neural networks; Closed-form solution; Feedback control; Frequency domain analysis; Mechatronics; Polynomials; Sampling methods; Shape control; Spline; Stability analysis; B-spline network (BSN); Learning feedforward control (LFFC); frequency domain analysis; high-order B-spline; iterative learning control (ILC);
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Mechatronics and Automation, Proceedings of the 2006 IEEE International Conference on
Conference_Location :
Luoyang, Henan
Print_ISBN :
1-4244-0465-7
Electronic_ISBN :
1-4244-0466-5
Type :
conf
DOI :
10.1109/ICMA.2006.257651
Filename :
4026163
Link To Document :
بازگشت