It has been shown that a driver assistance system using foresight information can reduce the fuel consumption with up to 15 %. Therefore information about e.g. road gradients, curves, intersections and speed limits is necessary. Currently available digital maps do not contain the required information, and the preparation and updating of such maps are associated with high costs. Instead of a map, a self-updating system based on appropriate vehicle sensors and a simple data storage facility is currently developed at the Institute of Internal Combustion Engines and Automotive Engineering (IVK) at the University of Stuttgart. The sensors of the system gather information about the road characteristics and this information is analysed with some added software algorithms for situation detection. The identified events are stored in a vehicle internal database and can in subsequent drives be used to improve the function of some efficiency, comfort or safety systems. Based on the fact that many vehicles repeatedly travel the same route, the validity of the identified events can be improved and updated by each passing of the event, i.e. a learning process. Possible fields of application are fuel efficient power train control, also improved curve speed warning and adaptive headlights as well as improved controlling of the energy flow in hybrid powered vehicles. The situations described in this paper can be identified by using only 'standard sensors' found in vehicles equipped with any form of an electronic stability system. Equipping the vehicle with more sensors, such as short and long range radar or cameras for lane detection and night-view, would open the possibility for the identification of even more situations relevant for forward-looking driving. For vehicle fleets, especially for long distance transport, a sharing system of the learnt information in the database would be useful.