Radial basis function neural network control of an XY micropositioning stage without exact dynamic model

Qingsong Xu, Yangmin Li

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

14 Citations (Scopus)

Abstract

In this paper, an adaptive neural sliding mode control based on radial basis function (RBF) neural network (NN) is implemented on a piezo-driven XY parallel micropositioning stage for a sub-micron accuracy motion tracking control. The controller is designed to map the relationship between the sliding surface variable and voltage applied to piezoelectric actuator (PZT). Hence, neither a hysteresis model nor an exact system dynamic model is required for the control purpose. The weight parameters of RBF NN are updated by an adaptive adjustment law via on-line learning. The effectiveness of the realized controller over traditional PID controller is demonstrated through experimental studies and the influences of design parameter variations on control performances are evaluated as well. Experimental results show that the intelligent controller can compensate for the hysteresis effectively and lead to a well-performance motion tracking within a specific input rate.
Original languageEnglish
Title of host publication2009 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2009
Pages498-503
Number of pages6
DOIs
Publication statusPublished - 4 Nov 2009
Externally publishedYes
Event2009 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2009 - Singapore, Singapore
Duration: 14 Jul 200917 Jul 2009

Conference

Conference2009 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2009
Country/TerritorySingapore
CitySingapore
Period14/07/0917/07/09

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Computer Science Applications
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Radial basis function neural network control of an XY micropositioning stage without exact dynamic model'. Together they form a unique fingerprint.

Cite this