A Robust Neural Network Controller
T.P. Leung, Qijie Zhou, Hailong Pei · 1992
In this paper we propose a new strategy for nonlinear system control based on the true inverse-dynamics learning. Variable structure control method is introduced to robustify the neural network controller. This scheme is applied to control a two-link robotic manipulator. The simulation results demonstrate that this scheme can achieve fast and precise robot motion control under the circumstances of load changing and inaccuracy of inverse-dynamics learning.