Methodology of evaluating safety in automobiles using intelligent sensor architecture and neural networks

This paper deals with the methodology of creating a new intelligent system to improve the driver's safety and comfort in an automobile transportation. A description of the necessary hardware and architecture is made in detail. Then the mathematical body of the system is described as well as its application on the acquired data. The application of the model on the data is made in two phases, first off-line and afterwards on real time acquired data. All the elaboration of the data results in one safety index for the current driver and situation. The data is always stored in the system's memory in order for the system to self-improve with time.