Gaze control for driver assistance systems

Currently available driver assistance systems more or less all rely on odometric data. However, in the future, a new form of information is likely to be used: visual sensors, e.g. cameras, which can provide additional knowledge on traffic and the vicinity. In this paper we present an inertially stabilized camera platform for high-precision tracking of vehicles, lane markers and traffic signs. A highly miniaturized inertial measurement system measures a vehicles motion in space and determines an effective stabilization command for the camera motion device. In addition, an image processing software extracts the position and self-motion of the points of interest. After presenting hardware and controller design, the paper focuses on the application in test drives on German freeways.