Vehicle fault detection and diagnosis combining an AANN and multiclass SVM

The main goals of a fault diagnosis system in a vehicle are to prevent dangerous situations for occupants. This domain is a complex system that turns the monitoring task a very challenging one. This paper presents a new approach based on history data process. In the first phase the approach learns behaviour from normal operation of the system using an autoassociative neural network. In the second phase a multiclass support vector machine classifies the type of fault present giving the final diagnosis. Results are shown for a ten variables vehicle monitoring.