Recurrent neural network control for nonlinear friction with sliding mode and friction estimator

Author

Han, Seong Ik ; Lee, Kwon Soon ; Yeo, Dae Yeon ; Koo, Kyung Wan ; Kim, Sae Han ; Ahn, Woo Young

Author_Institution

Dept. of Electr. Eng., Dong-A Univ., Busan, South Korea

Volume

3

fYear

2010

fDate

10-12 Aug. 2010

Firstpage

1387

Lastpage

1392

Abstract

In this paper, we have developed an friction compensation scheme for the rotary servo system using sliding mode and recurrent neural networks with friction estimator. An adaptive parameter updating rules of the recurrent neural network (RNN) and friction estimator have been developed to mimic the ideal sliding mode control and compensate the nonlinear friction via the Lyapunov stability theory.

Keywords

Lyapunov methods; friction; neurocontrollers; nonlinear control systems; recurrent neural nets; servomechanisms; variable structure systems; Lyapunov stability; friction compensation; friction estimator; nonlinear friction; recurrent neural network control; rotary servo system; sliding mode; Approximation methods; Artificial neural networks; Friction; Recurrent neural networks; Servomotors; Uncertainty; Dynamic friction; LuGre friction mode; Recurrent neural network; Servo system; Sliding mode control; component;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation (ICNC), 2010 Sixth International Conference on

Conference_Location

Yantai, Shandong

Print_ISBN

978-1-4244-5958-2

Type

conf

DOI

10.1109/ICNC.2010.5583904

Filename

5583904