Adaptive Neural Network Control for Multiaxis Trajectory Tracking: Nanomanipulation Example

Abstract

This paper investigates adaptive neural network control for multiaxis trajectory tracking of a piezoelectric actuator-driven nanomanipulation system. An approximation model-based control scheme which is involved with only nominal parts of the unknown system dynamics is designed first and the hysteretic effect is compensated by designing a disturbance observer. Then, adaptive neural networks are applied to approximate unknown parts and an adaptive neural network control scheme is designed. It can be proved that all the error signals are ultimately bounded with Lyapunov’s stability theory. A nanomanipulation experiment is conducted and the results prove the effectiveness of the proposed control.