Abstract
In this article, we develop a hybrid control scheme of a dynamic structured learning recurrent neural network (DRNN) with friction observer. The DRNN controller with the adaptive dynamic friction observer based on the LuGre friction can compensate both the nonlinear friction and the uncertainty in order to improve the positioning performance of the mechanical servo system. A proposed control scheme is validated via simulation of the position tracking control of the servo system
Keywords: Dynamic structured recurrent neural network, LuGre friction, Dynamic friction observer, Rotary servo system