Processing signals from an inertial sensor using neural networks


Pham D., SAĞIROĞLU Ş.

INTERNATIONAL JOURNAL OF MACHINE TOOLS & MANUFACTURE, vol.36, no.11, pp.1291-1306, 1996 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 36 Issue: 11
  • Publication Date: 1996
  • Doi Number: 10.1016/0890-6955(95)00087-9
  • Title of Journal : INTERNATIONAL JOURNAL OF MACHINE TOOLS & MANUFACTURE
  • Page Numbers: pp.1291-1306

Abstract

An inertial sensor is a device for determining the location of a part by measuring parameters related to its inertia. This paper describes a neural-network-based method for processing signals from an inertial sensor to compute the orientation of a part. This method involves training a multi-layer perceptron (the most commonly applied neural network type) using the backpropagation algorithm to model the operation of the sensor by mapping its natural frequency of vibration to part location information. The paper details the experimental procedure for acquiring data to train and test different neural network models and the results obtained. The latter show that the proposed method yields orientation values with a comparable degree of accuracy to existing techniques while requiring much less computational effort. Copyright (C) 1996 Elsevier Science Ltd