An adaptive controller using wavelet network for five-bar manipulators with deadzone inputs

Abstract

In this paper, an adaptive model-based control scheme is proposed for tracking control of five-bar manipulators with deadzone inputs. The proposed controller is based on the combination of nominal dynamic model of the five-bar manipulator, a wavelet network and a deadzone precompensator. The wavelet network and the precompensator are used for compensating the unknown deadzone inputs, modeling errors and uncertainties of the five-bar manipulator. The adaptation laws are derived for tuning parameters of the precompensator and wavelet network. The efficiency of the proposed control scheme is verified by comparative simulations. © Springer-Verlag Berlin Heidelberg 2013.