Sensor fusion for improved vision based lane recognition and object tracking with range-finders

Lane recognition based on computer vision and autonomous intelligent cruise control (AICC) using range-finders like infrared laser or radar were developed independently in the past. Both systems are now close to market. Nevertheless, they still have problems at larger look ahead distances which cannot be solved by computer vision or AICC sensors solely. Their combination, however, allows to improve both functions at the same time. Simple attempts to do this proved not to be satisfying. This paper presents a sensor fusion approach using Kalman filter that exploits the available sensory information in an optimal manner. The potential found in theoretical investigations has been confirmed by successful practical tests.