Skip to Main content Skip to Navigation
Theses

Etude de la robustesse du contrôle intelligent face aux fautes induites par les radiations

Abstract : The so-called intelligent control techniques, such as Artificial Neural Networks and Fuzzy Logic, are considered as being potentially robust. Their digital implementation gives compact and powerful solutions to some problems difficult to be tackled by classical techniques. Such approaches might be used for applications working in harsh environment (nuclear and space).
The aim of this thesis is to study the robustness of artificial neural networks and fuzzy logic against Single Event Upset faults, in order to evaluate their viability and their efficiency for onboard spacecraft processes. A set of experiments have been performed on a neural network and a fuzzy controller, both implementing real space applications: texture analysis from satellite images and wheel control of a martian rover. An original method, allowing to increase the recognition rate of any artificial neural network has been developed and used on the studied network. Digital architectures implementing the two studied techniques in this thesis, have been boarded on two scientific satellites. One is in flight since one year, the other will be launched in the end of 1999.
Obtained results, both from software simulations, hardware fault injections or particle accelerator tests, show that intelligent control techniques have a significant robustness against Single Event Upset faults. Data issued from the flight experiment confirm these properties, showing that some onboard spacecraft processes can be reliably executed by digital artificial neural networks.
Complete list of metadatas

Cited literature [109 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00002966
Contributor : Lucie Torella <>
Submitted on : Tuesday, June 10, 2003 - 5:11:12 PM
Last modification on : Friday, December 11, 2020 - 8:28:03 AM
Long-term archiving on: : Friday, April 2, 2010 - 7:12:49 PM

Identifiers

  • HAL Id : tel-00002966, version 1

Collections

TIMA | CNRS | UGA

Citation

P. Cheynet. Etude de la robustesse du contrôle intelligent face aux fautes induites par les radiations. Micro et nanotechnologies/Microélectronique. Institut National Polytechnique de Grenoble - INPG, 1999. Français. ⟨tel-00002966⟩

Share

Metrics

Record views

283

Files downloads

802