Fuzzy Neural Network Control for a Single Flexible-Link Manipulator

Article type: Research Article

Authors: Tzes, Anthony | Peng, Pei-Yuan | Khorrami, Farshad

Affiliations: Polytechnic University, Mechanical Eng. Dept., 333 Jay Street, Brooklyn, NY 11201 | Polytechnic University, Electrical Eng. Dept., 333 Jay Street, Brooklyn, NY 11201

Abstract: In this article, a fuzzy neural network controller for a single flexible-link manipulator is considered. A backpropagation neural network operating in the specialized learning mode is employed to decrease the effects of the inherent system nonlinearities, like the motor static friction and the saturation of the electronic amplifier. The neural network output resembles that of a Pulse Width Modulated controller. A fuzzy cell space controller supervises the overall scheme and reduces the amplitude and repetitions of control switchings. The fuzzy controller rules are extracted from a rulebase parameterized in terms of the l2 and l∞ norms of the output error. Simulation studies are presented to indicate the effectiveness of the proposed algorithm.

DOI: 10.3233/IFS-1993-1407

Journal: Journal of Intelligent and Fuzzy Systems, vol. 1, no. 4, pp. 319-334, 1993