In order to improve road safety conditions, the European Union has set itself the objective of reducing the number of road deaths. The European ASSET project aims to improve the safety and protection of drivers, rolling stock and road infrastructure. Our thesis work is fully in line with the ASSET project and aims to develop theoretical and practical tools integrating a lambda vehicle for improving road safety. The objective of the thesis is to develop or use field communication buses within the vehicle, allowing the vehicle to communicate with all its components and collect important information that will then be communicated wirelessly to other vehicles (or to a central office) to provide information on the condition of the vehicle or road. We propose an intelligent ubiquitous system called PCAN (Pervasif Controller Area Network), which aims to be an automatic and intelligent embedded system in a vehicle to improve road safety and driving comfort. In practical terms, this PCAN system makes it possible to collect contextual information on the vehicle's external and internal environment and also to communicate wirelessly with other external systems. The system consists of an open CAN bus, a wireless communication system and an embedded computer. The open CAN bus consists of nodes that allow the direct addition of different sensors or actuators. Two different embedded computers were used: The CompactRio and a laptop computer. Thanks to the latter, wireless communication modules were used to exchange data via Wi-Fi, ZigBee and GPRS. All the information collected can be displayed to the driver via a touch screen and the PCAN system is implemented in a Citroën C4 vehicle. On the basis of this intelligent vehicle, we have developed scenarios and applications to improve road safety and driving comfort. An application of the ABS warning is developed and tested experimentally for two scenarios: a direct V2V warning and a V2I or indirect V2V warning. Concerning the V2V warning in ZigBee communication, an application of vehicle location in distress is proposed thanks to a mathematical modeling of the distance as a function of the signal quality. In addition, an optimization algorithm is used to search for the optimal filling station (or charging point in the case of an electric vehicle) according to different criteria. PCAN can collect contextual information about the vehicle and its environment, it can also establish wireless communications. These functionalities are enhanced by the possibility of carrying out the control/command of certain vehicle components connected to PCAN. We use the PCAN system to remotely control and command a mechatronic actuator in the vehicle. The control method is based on an original association of a PID with a Fuzzy Logic (LF) controller.Translated with www.DeepL.com/Translator
