基于模糊贝叶斯网的空战传感器资源管理方法  

Aircombat Sensor Resource Management Based on Fuzzy Bayesian Networks

作　　者：梅风华[1] 何鹏[2,3] 

机构地区：[1]海军装备研究院,上海200436 [2]光电控制技术重点实验室,河南洛阳471009 [3]中航工业洛阳电光设备研究所,河南洛阳471009

出　　处：《航空计算技术》2017年第1期35-38,共4页Aeronautical Computing Technique

摘　　要：空战中目标状态信息的不确定性、目标与我机相对态势是影响机载传感器资源分配问题的重要影响因素。针对此问题,提出一种基于模糊贝叶斯网(FBN)的空战传感器资源管理方法,以空战传感器资源管理中涉及的影响因素因果关系作为建网依据,将目标信息增量、目标威胁、飞行员指令作为证据变量驱动网络进行概率推理,从而获取空战传感器资源的分配结果。仿真结果表明,与传统方法相比,方法的自适应变间隔采样策略能够根据目标威胁及飞行员指令影响,管理空战态势不同阶段的传感器资源以满足空战作战任务需求。Aircombat sensor resource allocation is influenced by the factors in aircombat which include uncertainty of target state,target threat,etc.In this paper,a method for aircombat sensor resource management based on fuzzy Bayesian networks（FBN） is presented.FBN method utilizes causalities in combat situation to found networks.Probabilistic reasoning among the networks is carried out by using the fuzzy value of target information gain,target threat level and pilot command.Simulation results indicate that FBN method,compared with information gain（IG） method,is able to allocate sensor resource reasonably according to its adaptive sampling strategy to meet the requirements of aircombat mission.

关 键 词：传感器管理 模糊贝叶斯网 信息增量 目标威胁 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]