Multi-sensors fusion approach for driver assistance systems

Our research enters into the design of perception devices for intelligent vehicles. This device is supposed to be a piece of new driving assistance systems aiming at increasing road safety. The European project PROMETHEUS, has emphasized the problem of the interaction between the driver and the vehicle via a driver assistance system. This project has been the precursor of several other projects linked to intelligent vehicles. Especially, the European project named CARSENSE involves 11 manufacturers and researchers from the car domain. The purpose of this project is to develop a multi-sensors device which would provide reliable data of the front vehicle environment. This system should help to design some safety assistance systems such an obstacle avoidance module or a semi-autonomous driving assistance at low speed. Our work takes a full part of designing and development, of the CARSENSE fusion system box. We are particularly interested in data fusion algorithms giving to the driver some accurate and above all reliable and pertinent information with, regard to the current driving situation. In order to deal with information reliability, we have designed a perception algorithm that combines some tools dealing simultaneously with the inaccuracy and uncertainty of dynamic environment representation