Feasibility study of robust neural network motion tracking control of piezoelectric actuation systems for micro/nano manipulation

Authors

Gursel Alici, University of WollongongFollow
Julian Smith
Hwee Choo Liaw
Bijan Shirinzadeh

RIS ID

22729

Publication Details

Liaw, H., Shirinzadeh, B., Alici, G. Smith, J. (2007). Feasibility study of robust neural network motion tracking control of piezoelectric actuation systems for micro/nano manipulation. The 13th International Conference on Advanced Robotics (pp. 1256-1261). Korea: Springer.

Abstract

This paper presents a robust neural network motiontracking control methodology for piezoelectric actuation systemsemployed in micro/nano manipulation. This control methodologyis proposed for tracking of desired motion trajectories in thepresence of unknown system parameters, non-linearities includingthe hysteresis effect, and external disturbances in the controlsystems. In this paper, the control methodology is establishedincluding the neural networks and a sliding scheme. Particularly,the radial basis function neural networks are chosen in this studyfor function approximations. Stability of the closed-loop systemsand convergence of the position and velocity tracking errors tozero are assured by the control methodology in the presence ofthe aforementioned conditions. Simulation results of the controlmethodology for tracking of a desired motion trajectory ispresented. With the capability of motion tracking, the proposedcontrol methodology can be utilised to realise high performancepiezoelectric actuated micro/nano manipulation systems.