Home // VEHICULAR 2014, The Third International Conference on Advances in Vehicular Systems, Technologies and Applications // View article

Vehicle Online Monitoring Sytem Based on Fuzzy Classifier

Authors:
Diana Maria Gomez Jaramillo
Claudia Victoria Isaza Narvaez

Keywords: Fuzzy classifier; on-board diagnostics; online monitoring.

Abstract:
In the automotive sector, electronic, mechanical, and software components have evolved significantly, resulting in increased complexity in vehicle fault diagnosis. The use of fuzzy classification techniques has been adapted for the online diagnosis of complex systems. In particular, Learning Algorithm for Multivariate Data Analysis (LAMDA) fuzzy classifier provides additional information through the Global Adequacy Degree (GAD) allowing to perform early preventive actions and supporting the operator in the decision-making process. This paper presents a car fault diagnosis system based on the LAMDA fuzzy classifier. The algorithm identifies, while the vehicle is in motion (online monitoring), the state of the vehicle, i.e., normal driving behavior, aggressive driving (driving behavior reflecting an impatient or angry driver) or mechanical failure. The implementation of the monitoring system implementation is performed in a midrange Renault vehicle. The algorithm achieves a 92.52% correct functional state identification with a low computational cost.

Pages: 33 to 38

Copyright: Copyright (c) IARIA, 2014

Publication date: June 22, 2014

Published in: conference

ISSN: 2327-2058

ISBN: 978-1-61208-348-3

Location: Seville, Spain

Dates: from June 22, 2014 to June 26, 2014