Adaptive NN control of dynamic systems with unknown dynamic friction

Abstract

In this paper, based on the dynamic LuGre friction model, adaptive NN controllers are presented by using neural networks to parameterize the unknown characteristic function α(x, ẋ) or the unknown dynamic friction bounding function respectively. Using Lyapunov synthesis, the adaptive control algorithms are designed to achieve globally asymptotic tracking of the desired trajectory and guarantee the boundedness of all the signals in the closed-loop. Intensive simulations are carried out to verify the effectiveness of the proposed methods.