A two-time-scale neural controller for the tracking control of rigid manipulators

Author

Cheng, Weiying ; Wen, John Ting-Yung

Author_Institution

Dept. of Electr. Comput. & Syst. Eng., Rensselaer Polytech. Inst., Troy, NY, USA

Volume

24

Issue

7

fYear

1994

fDate

7/1/1994 12:00:00 AM

Firstpage

991

Lastpage

1000

Abstract

In this paper, a two-time-scale neural controller applied to tile tracking control of rigid manipulators is introduced. Several fast learning rules and slow learning strategies are proposed. The stability properties of the closed loop system with the proposed two-time-scale neural controller are analyzed. The results show that the tracking error will be uniformly bounded and converge to a bounded region. If a sufficiently large leakage term is used in the fast learning rule, the ultimate bound of the tracking error depends only on the accuracy of the slow learning. Moreover, the feasibility of the proposed neural controller is demonstrated through the simulation of a two-link rigid robot manipulator

Keywords

intelligent control; learning systems; manipulators; neural nets; position control; stability; closed loop system; fast learning rules; rigid manipulators; robot; slow learning strategies; tracking control; tracking error; two-time-scale neural controller; Closed loop systems; Computer errors; Error correction; Feedforward neural networks; Manipulators; Neural networks; Nonlinear dynamical systems; Nonlinear systems; Robots; Stability analysis;

fLanguage

English

Journal_Title

Systems, Man and Cybernetics, IEEE Transactions on

Publisher

ieee

ISSN

0018-9472

Type

jour

DOI

10.1109/21.297788

Filename

297788