Application of piezoelectric gyro's drift compensation algorithm based on neural network

Yu Liu*, Qiujun Li, Jun Liu, Leilei Li, Youju Mao

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

Time serial model (ARMA) and the neural network model based on single temperature input cannot describe well the behaviors of piezoelectric gyro's null drift. So the angle measurement error will not be compensated effectively. A new model based on the three layer error compensated BP neural network considering the temperature and run time input was proposed. The experiment data show that mean variance of the piezoelectric gyro's null drift error is decreased to 0.0128. It is only 8.42% of uncompensated value. Mean variance of scale factor is decreased to 1.19 × 10-6. This result is only 33.3% of uncompensated value. And the practicability of this model was proved by the practical measurement.

Original languageEnglish
Title of host publicationProceedings of the World Congress on Intelligent Control and Automation (WCICA)
Pages4823-4826
Number of pages4
DOIs
Publication statusPublished - 2006
Externally publishedYes
Event6th World Congress on Intelligent Control and Automation, WCICA 2006 - Dalian, China
Duration: 21 Jun 200623 Jun 2006

Publication series

NameProceedings of the World Congress on Intelligent Control and Automation (WCICA)
Volume1

Conference

Conference6th World Congress on Intelligent Control and Automation, WCICA 2006
Country/TerritoryChina
CityDalian
Period21/06/0623/06/06

Keywords

  • Compensation
  • Drift
  • Neural network
  • Piezoelectric gyroscope
  • Scale factor

Fingerprint

Dive into the research topics of 'Application of piezoelectric gyro's drift compensation algorithm based on neural network'. Together they form a unique fingerprint.

Cite this