态势估计中的目标分群算法设计与实现

Design and Realization of Target Grouping in Situation Assessment

针对态势估计中的目标分群问题,分析了目标的位置、运动状态和属性对分群结果的影响,建立了目标相似度计算模型,并提出了分群策略,包括群的形成、分裂和合并等。采用面向对象软件设计思想建立了分群对象模型,分析了对象间关系。给出了分群算法的主体流程,并对其中的关键步骤进行了详细的说明。开发了目标分群算法模块,实现了对数据的预处理、目标群的更新维护、群属性的计算等功能,通过仿真试验,调整目标分群影响因子及加权因子,得到了较好的分群效果,验证了文中方法的可行性和正确性。

Situation assessment is one component of data fusion.And target grouping is one of the main tasks of situation assessment.The factors of target,such as position and attribute,which influence the result of grouping,are analyzed. The grouping strategy is also presented for initialization,split,and combination.We used object-oriented technology to build grouping object model.The algorithms are brought forward for calculating target grouping. Particularly,the key steps in algorithms are described. We developed grouping dynamic link library program. The program can provide pretreatment of data,group updating,and group attribute calculation. The algorithm parameters and weighting factors are adjusted by simulation experimentation. Simulation results validate the feasibility and correctness of the grouping algorithms.