一种基于差异熵赋权法的无人机能力评估方法
An UAV Capability Evaluation Method based on Differential Entropy Weighting
王岚清 1杨刚 1原郭文1
作者信息
- 1. 西北工业大学计算机学院,西安,710129
- 折叠
摘要
在评估无人机性能方面,为有效克服主观性强、对数据过于敏感、计算复杂度高等缺陷,提出无人机"能力"建模方法.该方法在向上层研究屏蔽无人机异构性和资源多样性的同时保留无人机的个性化重要特征,呈现无人机"能力"以支持上层模型设计和算法设计,从而减轻了上层研究的先验工作量,推进了相关研究的开展.进一步提出基于无人机"能力"的差异熵赋权法,该方法依据判断矩阵最大特征值所对应的特征向量确定各个特性的相对权重,加权计算得到最终的评估结果,从而有效地对无人机各能力以及综合性能进行相对评估.试验结果表明:差异熵赋权法在克服主观偏差的同时,增加了指标间横向对比,提升了计算稳定性.
Abstract
In the aspect of evaluating UAV performance,in order to improve the disadvantages,such as strong subjectivity from the experts,too sensitive to data,the UAV capability modeling is proposed.It not only shields the heterogeneity and resource diversity of UAV from the upper layer,but also retains the important characteristics of UAV,which can present the UAV's capability to support the design of upper layer model and algorithm,and reduce the prior workload of the upper-level research and promote the development of related research.Furthermore,the differential entropy weighting method based on UAV capability is also proposed to evaluate the capabilities and comprehensive performance of UAV effectively.This method determines relative weight of each feature according to the feature vector corresponding to the maximum eigenvalue of the judgment matrix,and the final evaluation result is obtained through weighted calculation.Experimental results show that the method overcomes subjective bias from experts and increases the horizontal comparison among indicators to improve the stability of calculation.
关键词
无人机/能力建模/差异熵赋权/层次分析法/熵权法Key words
UAV/capability modeling/differential entropy weighting/AHP/entropy weighting引用本文复制引用
基金项目
国家重点研发计划(2017YFB1001900)
出版年
2024
NETL
NSTL
万方数据