关键词: 数据关联, 群目标跟踪, 多假设跟踪（MHT）, 群目标质心跟踪

Abstract: Considering the limitation of the traditional group-targets centroid tracking methods in tracking individual targets exactly, a new group tracking algorithm is proposed based on the Multiple Hypothesis Tracking（MHT） concept. The algorithm is different with general centroid group tracking method. It not only maintains the groups, but also tracks the individual targets by using MHT to deal with complicated data association problem. The extreme situations, for example, when ECM and chaffs exist, are taken into account to prove tracking stability. Finally, plentiful simulations are carried out to evaluate the algorithm’s performance and the simulation results indicate that the closely spaced objects in group are accurately tracked by the proposed algorithm and the tracking maintain stability when in the clutter environment which the clutter density is 1×10－6 per square meter.

Key words: data association, group tracking, Multiple Hypothesis Tracking（MHT）, group-targets centroid tracking