摘要
联合概率数据关联算法(JPDA)是密集杂波环境下一种有效的多目标跟踪算法,但该算法的复杂度会随着目标和观测值的增加而显著增长。为了减少JPDA算法所需要的存贮空间和计算时间,提出了一种改进的联合概率数据关联算法(I-JPDA)。首先通过合理选取跟踪门门限的阈值,去除小概率事件,然后再根据跟踪门内目标的关联概率对关联事件的概率密度值进行衰减,计算出跟踪门内各目标的关联概率。经过理论分析和仿真试验,证明了该方法在保证跟踪成功率的同时,还具有算法简单、计算量小和易于工程实现等优点。
Joint Probabilistic Data Association（JPDA） is an effective algorithm of multiple targets tracking in dense clutter.But its complexity is outstandingly increased with increased targets and measurements.To reduce expense of memory and time for JPDA,a new improved JPDA algorithm was introduced for data association.Firstly,in this new method,removing the small probability events by selecting the right threshold values of the targeting gate;secondly,fade the probability density of association events according to the association probability of the targets in the targeting gate,after that compute the association probability of the targets.The theory analysis and computer simulation show that this algorithm can keep the rate of success of the target tracking with low CPU computing time and easy come true in engineering application.
出处
《火力与指挥控制》
CSCD
北大核心
2010年第4期106-110,共5页
Fire Control & Command Control
关键词
目标跟踪
联合概率数据关联
关联概率
系统仿真
target tracking
joint probabilistic data association
association probability
system simulation
