Multi-Mems Differential Pressure Sensor Elements-Based Airflow Sensor with Neural Network Model

Kotaro Haneda, Kenei Matsudaira, Hidetoshi Takahashi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

This paper reports a compact spherical airflow sensor using multi-MEMS differential pressure (DP) sensor elements. Three built-in MEMS sensors simultaneously measure the DP around the spherical housing structure so that the measured DPs are converted into 2D wind direction and speed. The sensor outputs are converted into wind direction and speed by neural network. We attached the calibrated sensor to a toy drone as a demonstration. Then, it was confirmed that the output corresponding to wind direction and speed was measured when a crosswind was applied during flight.

Original languageEnglish
Title of host publication2023 IEEE 36th International Conference on Micro Electro Mechanical Systems, MEMS 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages499-502
Number of pages4
ISBN (Electronic)9781665493086
DOIs
Publication statusPublished - 2023
Event36th IEEE International Conference on Micro Electro Mechanical Systems, MEMS 2023 - Munich, Germany
Duration: 2023 Jan 152023 Jan 19

Publication series

NameProceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS)
Volume2023-January
ISSN (Print)1084-6999

Conference

Conference36th IEEE International Conference on Micro Electro Mechanical Systems, MEMS 2023
Country/TerritoryGermany
CityMunich
Period23/1/1523/1/19

Keywords

  • Airflow sensor
  • Differential pressure sensor
  • Machine learning
  • Neural network

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Mechanical Engineering
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Multi-Mems Differential Pressure Sensor Elements-Based Airflow Sensor with Neural Network Model'. Together they form a unique fingerprint.

Cite this