Advanced Filtering Techniques for Multisensor Vehicle Tracking
Eric Richter, Robin Schubert, Gerd Wanielik · 15th World Congress on Intelligent Transport Systems and ITS America's 2008 Annual MeetingITS AmericaERTICOITS JapanTransCore · 2008
The robust and reliable detection of objects in the path of a vehicle is an important prerequisite for collision avoidance and collision mitigation systems. In this paper, the authors present an ego-motion compensated tracking approach that combines radar observations with the results of a contour-based image processing algorithm. This approach is able to handle all uncertainties of the system in a unified way without analytical linearization by using the Unscented transform. With that approach, the covariances of the system can be estimated more accurately. The authors describe both the image processing and the state estimation algorithms, the present the results of several practical tests. The system is able to fuse the data from radar and image sensors to estimate the position, direction, and width of objects in front of the vehicle. Other vehicles can also be detected by this image processing system.