摘要
最近邻域经典算法在求解航迹关联问题时,由于过度依赖特征阈值以及缺乏全局性考虑,在航迹密度较高的情况下容易出现错误关联.针对这一问题,引入全局搜索策略并采用动态规划和跟踪门技术,提出了一种新的全局最优航迹关联算法.在真实的海上目标航迹关联环境下对两种算法进行了实现,与最近邻域算法相比,新算法不仅获得了较高的关联正确率,同时减少了关联结果对特征阈值的依赖.
Because of being highly dependent on the threshold and lacking the consideration of global solutions,the Nearest Neighbors Algorithm(NNA) will make some mistakes when the density of targets is high.To solve this problem,global search strategy,gating and dynamic programming are used to build a new correlation algorithm-the Global Best Track Correlation Algorithm(GBTCA).In two experiments,both of the two algorithms are run and compared to NNA,GBTCA,which shows that the new algorithm has a higher correct correlation rate and is less dependent on threshold values.
出处
《西安电子科技大学学报》
EI
CAS
CSCD
北大核心
2012年第1期67-74,共8页
Journal of Xidian University
基金
陕西省自然科学基础研究计划资助项目(2010JM8027)
关键词
数据融合
航迹关联
全局搜索
动态规划
跟踪门
data fusion
track correlation
global search
dynamic programming
gating