Tracking vehicles using radar detections

This paper is concerned with the problem of tracking vehicles using radar detections. In particular, we deal with problems where multiple detections are received from each target. According to recent studies, automotive radar sensors often receive detections from a discrete set of reflection centers. From these results, we develop a family of sensor models suitable for tracking. A significant difficulty in performing tracking using models from this family is that it involves solving a data association problem with a large number of association hypotheses. To reduce the complexity of this problem, we propose a framework where similar hypotheses are joined into groups. Following this approximation, basic data association algorithms can be implemented and initial studies using both simulated data and real measurements show a promising performance.