基于模糊动态贝叶斯网空战敌方作战企图识别  

Foe intention inference in air combat based on fuzzy dynamic Bayesian network

作　　者：余振翔[1,2] 胡笑旋[1,2] 夏维[1,2] 

机构地区：[1]合肥工业大学管理学院,安徽合肥230009 [2]合肥工业大学飞行器网络系统研究所,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2013年第10期1210-1216,1253,共8页Journal of Hefei University of Technology：Natural Science

基　　金：中航工业产学研合作创新工程专项资助项目CXY2011HFGD20);中央高校基本科研业务费专项资金资助项目(2012HGZY0009)

摘　　要：文章对空战中敌方目标作战企图的各个因素进行了分析,建立了敌方目标作战企图识别的动态贝叶斯网模型,根据连续变量的不同特点,通过构造和设置不同的隶属函数和参数,对连续变量的观测值进行模糊分类,通过引入可能性概率转换理论转换为贝叶斯网能够利用的证据信息用于推理,使对敌方目标作战企图的推理过程具有前后连续性,更加符合客观实际,有效地减少了主观判断带来的误差。仿真实验表明,该方法的推理结果与理论分析一致,为识别敌方目标作战企图提供了一种可行且有效的方法。The dynamic Bayesian network model for foe intention inference is established based on the analysis of the characteristics concerning enemy target combat intention. According to different char- acteristics of continuous variable, different membership functions and parameters are constructed and set to classify continuous variables, and then the membership degree is converted into the evidence which can be used to Bayesian network inference by probability-possibility transformation theory. On the one hand, this inference method has continuity which is more suitable to real situation. On the other hand, this method reduces errors caused by subjective judgment. The simulation results show that the inference results of this method are consistent with those of the theoretical analysis, which proves that it is a feasible and effective method for foe intention inference.

关 键 词：空战 作战企图识别 动态贝叶斯网络 模糊集合 推理 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]