基于复杂网络和朴素贝叶斯的飞行态势评估

Flight situation assessment based on complex networks and naive Bayes

为了辅助管制员作出合理的冲突调配决策,准确地评估空中飞行态势,提出了一种基于复杂网络和朴素贝叶斯分类器的飞行态势评估方法。首先,依据航空器的位置、航速、航向等重要信息,通过三维速度障碍法构建了冲突网络,在此基础上选取了平均度、平均点强、平均加权聚类系数等6个网络特征指标构建态势评估指标体系。采用分类的思想对飞行态势进行评估,分类器选择朴素贝叶斯分类器,将飞行态势分为优、良、中、差4个等级。仿真试验结果表明:相对于传统飞行状态网络,所构建的冲突网络能够降低33.3%的虚警率;所提出的评估模型能够准确判断空中的飞行态势,能为航空管制工作提供辅助决策。

In order to assist controllers to make reasonable conflict deployment decisions, and to accurately evaluate the in-flight state, a flight situation evaluation method using the complex network and naive Bayes classifier(NBC) was proposed. Firstly, based on the important information such as the aircraft position, speed and heading, a conflict network was constructed with the three-dimensional velocity obstacle method. Six network features including the average degree, average node strength and average weighted clustering coefficient were selected to build the situation evaluation indicator system. The classification idea was employed to evaluate the flight situation. Meanwhile, NBC was selected as the classifier. The flight situation was classified into 4 grades(excellent, good, medium and poor). The results of the simulation experiments show that the constructed conflict network can reduce the false alarm rate by 33.3% compared with the traditional flight status network. Moreover, the proposed evaluation model can accurately determine the flight situation in the air and support the aviation control work.