Vibration suppression and tracking control of a flexure-jointed motion stage mechanism using LTV-FIR command filtering

Research output: Chapter in Book/Report/Conference proceedingsChapterpeer-review

Abstract

This paper presents a linear time-varying (LTV) prefilter design for residual vibration suppression in LTV motion systems. As tracking performance can be affected by delay effects from prefiltering, an additional moment constraint is introduced in the prefilter synthesis equations for tracking of constant velocity commands. Inclusion of the moment constraint can result in higher prefilter gain (H2 norm) which leads to overshoot in the transient response. It is shown how this effect can be reduced by noncausal prefiltering with specified preview time. The LTV-FIR prefilter is applied experimentally to a multi-flexure X-Y motion stage mechanism where the position-dependent dynamics are captured by interpolation of a set of identified LTI models. The experimental results confirm that the prefilter with input preview can reduce settling time and overshoot compared with the causal prefilter, while maintaining good tracking performance.

Original languageEnglish
Title of host publication2021 IEEE International Conference on Mechatronics, ICM 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728144429
DOIs
Publication statusPublished - 7 Mar 2021
Externally publishedYes
Event2021 IEEE International Conference on Mechatronics, ICM 2021 - Kashiwa, Japan
Duration: 7 Mar 20219 Mar 2021

Publication series

Name2021 IEEE International Conference on Mechatronics, ICM 2021

Conference

Conference2021 IEEE International Conference on Mechatronics, ICM 2021
Country/TerritoryJapan
CityKashiwa
Period7/03/219/03/21

Keywords

  • Filtering
  • Linear time varying system
  • Motion control
  • Vibration

Fingerprint

Dive into the research topics of 'Vibration suppression and tracking control of a flexure-jointed motion stage mechanism using LTV-FIR command filtering'. Together they form a unique fingerprint.

Cite this