计算机工程与应用 ›› 2021, Vol. 57 ›› Issue (24): 100-106.DOI: 10.3778/j.issn.1002-8331.2012-0208

• 理论与研发 • 上一篇    下一篇

基于改进修正权重的证据组合方法

蒋斌,梁小安,张亮,高杨军   

  1. 空军工程大学 装备管理与无人机工程学院,西安 710051
  • 出版日期:2021-12-15 发布日期:2021-12-13

Evidence Combination Method Based on Improved Modified Weight

JIANG Bin, LIANG Xiao’an, ZHANG Liang, GAO Yangjun   

  1. College of Equipment Management and Unmanned Aerial Vehicle Engineering, Air Force Engineering University, Xi’an 710051, China
  • Online:2021-12-15 Published:2021-12-13

摘要:

针对证据组合过程中修正权重的确定问题,提出了一种综合考虑证据间关联关系和自身不确定度的改进修正权重确定方法,实现了对待组合证据的有效修正。基于对证据信息随机性和不具体性的阐述,得到了证据综合不确定度的度量方法。结合基于改进冲突度的相似性权重和基于综合不确定度的可靠性权重,确定了改进的修正权重。应用Dempster组合规则对修正后的加权平均证据进行有效组合。算例分析结果表明,所提方法能够充分利用证据信息,解决证据冲突,组合结果更为客观、准确,可信度更高。

关键词: 证据理论, 修正权重, 不确定度, 相似性, 可靠性

Abstract:

For the problem of determining the modified weight in the process of evidence combination, a method of determining the improved modified weight is proposed, which takes into account the correlation and the uncertainty degree of the evidence. The measurement method of comprehensive uncertainty degree of evidence is obtained based on the elaboration of randomness and non-concreteness of evidence information. The improved modified weight is determined by combining the similarity weight based on improved conflict degree and the reliability weight based on comprehensive uncertainty degree. Dempster’s rule of combination is applied to combine the modified weighted average evidence effectively. The results of numerical example analysis show that the proposed method can make full use of evidence information, solve evidence conflicts, and the combined results are more objective, accurate and reliable.

Key words: evidence theory, modified weight, uncertainty degree, similarity, reliability