基于改进证据网络的空战动态态势估计方法  

Dynamic situation assessment method of aerial warfare based on improved evidence network

作　　者：王昱[1,2] 章卫国[1] 傅莉[2] 黄得刚 李勇[3] 

机构地区：[1]西北工业大学自动化学院,西安710129 [2]沈阳航空航天大学自动化学院,沈阳110136 [3]沈阳工业大学电气工程学院,沈阳110870

出　　处：《航空学报》2015年第12期3896-3909,共14页Acta Aeronautica et Astronautica Sinica

基　　金：国家自然科学基金(61374032);航空科学基金(2012ZA01011);辽宁省教育厅科学研究一般项目(L2015412)~~

摘　　要：针对无人机空战态势估计既需要综合考虑多类型因素影响,又需要具有不确定性推理能力的特点,建立了一种基于改进证据网络的动态态势估计模型并设计了威胁等级评估推理方法。首先,考虑空战决策时间较短的特点,提出变量框架等级缩减方法以提高网络运行效率;然后,针对空战态势信息具有大不确定性的特点增加冲突数据自适应融合算法以及网络证据的时间序列预测,提高证据的合理性;最后,引入时空融合思想和变权机制将前一时刻的威胁信息作为评判后一时刻威胁的重要标准,应用威胁在时间维度上的递归合成增加信息在时间方向的传递,改善了由于证据失真引起的评估结果的不合理性问题,并通过仿真验证了该方法的有效性。Aimed at the comprehensive consideration for the impact types and the need of reasoning ability with uncertainty that situation assessment of UAV aerial warfare requires,a model of dynamic situation assessment based on improved evidence network is established and the threat level evaluation reasoning method is designed.Firstly,considering the characteristics of short decision time,agrade reduction method for variable recognition frame is proposed to improve the network operation efficiency.Then according to the characteristics of large uncertainties of aerial warfare situation information,the adaptive combination arithmetic of conflicting data and the time series prediction for evidence are added to advance the rationality of evidence.Finally,the temporal-spatial fusion thought and the variable weight mechanism are also introduced to make the threat information of previous time an important standard for the threat assessment of the next time.Due to the threat recursion combination in the temporal direction,the threat information transmission is increased.It is verified that the problem of irrational assessment results caused by the distortion of information is improved and the proposed method is effectively validated by the simulation examples.

关 键 词：改进证据网络 态势估计 推理 等级缩减 证据合理性 时空信息融合 变权 

分 类 号：V247.1[航空宇航科学与技术—飞行器设计] TP181[自动化与计算机技术—控制理论与控制工程]