Abstract
This paper presents an adaptive neural-fuzzy control scheme for a dual-level-structure flexible manipulator with variable dynamic payload. The dynamic moving model of the flexible manipulator is derived and the state-space equation is formulated first. A control scheme that consists of a neural-fuzzy controller in the feedback channel and an image-guided identification network (IGIN) in the forward configuration is then proposed. The IGIN is employed to locate the object (e.g. bimetal) to achieve the tracking function, while the dynamic neural network is used to learn the weighting factor of the fuzzy controller. Finally, simulations are run for various modes to describe the dynamic tracking system, and simulated results show a good performance of the control tracking system.
Get full access to this article
View all access options for this article.
