An Improved Version of a Multiple Target Tracking Algorithm
Publication Type:
Journal ArticleSource:
Information & Security: An International Journal, Volume 9, p.130-140 (2002)Keywords:
assignments, multiple hypotheses trackingAbstract:
The author presents an improved version of an algorithm for multitarget tracking. The algorithm extends the seminal algorithm of Nagarajan et. al., overcoming to some extent combinatorial problems, arising while simultaneously tracking multiple targets using track-while-scan radars. In a previous work, the algorithm was tested in comparatively simple to moderate scenarios. In this article, the author changes some of the previously suggested additional rules of the Nagarajan’s algorithm and presents more comprehensive experimental results with more sophisticated scenarios, using at the same time more processing power. By an additional experiment, the author reveals a useful feature of the presented algorithm that makes it a viable alternative to the well known algorithms for finding the first K-best hypotheses in the framework of the MHT approach.