A fixed size assumption based data association method for coalescing objects tracking using a laser scanner sensor

Data association is one of the crucial parts in the reliable objects' tracking system. In this paper, a methodology for data association using an Oriented Bounding Box based object representation is presented. A laser scanner sensor is used for objects perception. A data association algorithm for coalescing objects is described. The algorithm is based on the Nearest Neighbours principle enriched by an Inter-Rays uncertainty paradigm and a Fixed Size assumption, which are introduced to improve the tracking process in terms of objects size and center position estimation. Experimental results are presented to demonstrate the effectiveness of the proposed technique.