摘要
为了克服杂波环境下对多目标进行数据互联时,计算量出现组合爆炸现象,提出了改进的基于FCM的多目标跟踪数据关联算法。将航迹的预测值转换到各个传感器的观测空间作为各自的聚类中心,利用目标属于所有量测的隶属度,来代替JPDAF中的关联概率,将多目标数据关联问题可转化为模糊聚类问题,进行关联计算。改进的基于FCM的多目标跟踪数据关联算法,有效地利用了目标状态估计中的历史信息,实现量测与航迹的关联。该算法克服了JP-DAF算法计算量大的缺点,实现杂波环境下多目标数据互联。仿真结果表明了该算法的有效性。
An improved FCM- based multi- target tracking data association algorithm （MFCMDA） is proposed to overcome the combinatorial explosion of calculating volumes when multi - objective data are on the Internet under the clutter environment. The forecast track would be converted into various sensors in space as their own observations of the cluster, and by substituting the membership measured by all quantities belong to the goal instead of associated probability in JPDAF, the multi - objective data associated problem can be converted into a fuzzy clustering one in performing associated computation. So the improved FCM based multi -target tracking data association algorithm can be taken to realize the association of track and measure by effectively using the historical information in target state estimation. The improved FCM based multi - target tracking data association algorithm is superior to the JPDAF algorithm in computation. The simulation results show that the method is effective.
出处
《空军工程大学学报（自然科学版）》
CSCD
北大核心
2010年第1期36-39,共4页
Journal of Air Force Engineering University(Natural Science Edition)
基金
航空科学基金资助项目(01I30011)
关键词
目标跟踪
数据关联
隶属度
模糊聚类
target tracking
data association
membership
fuzzy clustering
