摘要
基于最大熵准则实现机动目标状态融合估计,针对聚类算法中隶属度计算与聚类中心耦合、类中心初始化不当的引入误差问题,提出向量解析方法解耦求解隶属度。仿真实验表明,该方法能避免聚类中心难选择问题,具有良好的估计精度。
A new approach to the state fusion estimation of moving targets based on the maximum entropy principle is described. Targeting at the fact that the membership computation is coupled with the cluster centre in the cluster algorithm and the errors caused by improper cluster centre initialization, a vector analytic method is proposed to solve the membership. Simulation results indicate that the method avoids embarrassment： in cluster centre selection and has a good estimation accuracy.
出处
《指挥信息系统与技术》
2011年第2期19-22,70,共5页
Command Information System and Technology
关键词
数据融合
最大熵
模糊聚类
data fusion
maximum entropy
fuzzy clustering
