Behavior-based learning of visual characteristics from real-world traffic scenes for driver assistance systems

摘要

The invention relates to method for setting-up and operating a driver assistance system which visually senses the environment of a vehicte, and senses at least one rate of change of a parameter representing a status of the car, which parameter can be influenced by a driver manipulating a man/machine interface of the vehicle. The rate of change is classified in e.g. braking, accelerating, turning left or right. The data stream of the visual sensor is split in categories, in line with the classification of the rate of change, providing a link between images and driver behavior. If a similar Image is recorded, driver behavior can be predicted.