A neural network a neural network approach for the design of micromachined accelerometers

Elena Gaura, N. C. Steele, R. J. Rider

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


The features which govern the development of a closed loop micromachined accelerometer using Neural Networks are presented Two modular Neural Networks have been designed, trained and incorporated into the closed loop transducer structure. a compensating Neural Network (which performs a static mapping) and a feedback Neural Network which has a twofold role. Firstly, it Linearises the feedback relationship between the system's output and the electrostatic forces acting on the electrodes. Secondly, the network demodulates the output signal in order to apply the feedback to only one electrode at a time. Simulation results and the advantages of this approach over the conventional design are shown.

Original languageEnglish
Title of host publicationComputational Intelligence for Modelling, Control and Automation
Subtitle of host publicationNeural Networks & Advanced Control Strategies
EditorsM Mohammadian
PublisherIOS Press
Number of pages4
ISBN (Print)9051994737
Publication statusPublished - Jan 1999
Externally publishedYes
EventInternational Conference on Computational Intelligence for Modelling, Control and Automation (CIMCA 99) - VIENNA, Austria
Duration: 17 Feb 199919 Feb 1999

Publication series

NameConcurrent Systems Engineering Series
PublisherIOS Press
ISSN (Print)1383-7575


ConferenceInternational Conference on Computational Intelligence for Modelling, Control and Automation (CIMCA 99)

Cite this