Method for detecting and confirming critical driving situations

The method involves monitoring an object i.e. preceding vehicle, in front of a vehicle by independent sensing units (10-1-10k) attached to the vehicle. Signals supplied by the sensing units are provided as sensor signals (f1-fk) in form of a discrete probability density function of a physical measured variable i.e. time span, over a defined interval. The sensor signals are combined in form of probability density function of an end signal (fz) with the measured variable. Final rating of an actual driving condition is conducted based on the combined end signal by a learning classifier (12). The learning classifier is used as an artificial neural network or a support vector machine. Independent claims are also included for the following: (1) a computer program comprising a program code for executing a method for detection and confirmation of critical driving conditions (2) a computer program product comprising a program code for executing a method for detection and confirmation of critical driving conditions (3) a device for detection and confirmation of critical driving conditions.