摘要
通过分析导弹目标识别问题的特殊性,设计了适用于导弹目标识别的动态贝叶斯网络。根据导弹目标的特点,选取了几种具有较好普适性的目标特征,并给出了每种目标特征的条件概率。通过分析导弹目标识别流程,提出了合理的贝叶斯网络模型。最后经过仿真试验,证明该模型相比单特征识别模型识别概率有明显提高,并且具备较好的稳健性。
A dynamic Bayesian network is constructed for missile target recognition after analysis of its particularity.In light of the features of missile targets,several universal target features are chosen and assigned appropriate conditional probability for the problem.Then,missile recognition flowchart is illustrated and a rationalized dynamic Bayesian network model is proposed.Simulation is performed to evaluate the performance of the model.Results show that the model＇s performance is much better and more robust than single feature recognition models.
出处
《飞行器测控学报》
2011年第2期70-74,共5页
Journal of Spacecraft TT&C Technology
关键词
导弹目标
弹头识别
动态贝叶斯网络
贝叶斯估计
Missile
Warhead Recognition
Dynamic Bayesian Network（DBN）
Bayesian Estimation
作者简介
王爽（1987-），男，硕士研究生，主要研究方向为导弹目标融合识别技术；E—mail：galoiscode@gmail．com
