Model-based decentralized embedded diagnosis inside vehicles: Application to Smart Distance Keeping function

In this paper, the deployment of a fault diagnosis strategy in the Smart Distance Keeping (SDK) system with a decentralized architecture is presented. The SDK system is an advanced version of the Adaptive Cruise Control (ACC) system, implemented in a Renault-Volvo Trucks vehicle. The main goal of this work is to analyze measurements, issued from the SDK elements, in order to detect, to localize and to identify some faults that may be produced. Our main contribution is the proposition of a decentralized approach permitting to carry out an on-line diagnosis without computing the global model and to deploy it on several control units. This paper explains the model-based decentralized solution and its application to the embedded diagnosis of the SDK system inside truck with five control units connected via a CAN-bus using "Hardware In the Loop" (HIL) technique. We also discuss the constraints that must be fulfilled.