摘要
Dempster-Shafer证据理论广泛应用于信息融合中,但是在证据高度冲突情况下基于经典D-S组合规则的融合结果存在问题。针对这一问题,提出了一种最优的证据组合方法。新方法首先引入了一个加权证据间距离函数,提出了一个全局距离的概念,依据全局距离最小的最优化准则求出各证据的权重,之后对系统中的证据加权平均,最后再利用D-S组合规则实现信息融合。目标识别算例仿真的结果表明,该方法性能优于现有的加权组合方法,在处理冲突数据时能够有效快速地识别出目标。
Dempster-Shafer（D-S） method has been used widely in all kinds of data fusion system, but it has difficulty in dealing with combining evidences with high degree of conflict. In order to solve the problem, an optimal method was proposed. First, a weighted evidence distance function between the weighted bodies of evidence was introduced, and then the concept of global distance was brought forward, and the optimal weight of every evidence body was acquired, accoroling to the principle that the global distance is minimal. Then oweraging weighted evidence in the system, the D-S combination was used to realize information fusion. The simulation result of example testifies that the performance of the proposed approach is better than the existing approaches and more efficient in target recognition.
出处
《电机与控制学报》
EI
CSCD
北大核心
2009年第A01期178-182,192,共6页
Electric Machines and Control
基金
国家自然科学基金资助项目(60874105
60904099)
新世纪优秀人才支持计划资助项目(NCET-08-0345)
上海市青年科技启明星计划资助项目(09QA1402900)
上海交通大学"晨星学者计划"资助项目(T241460612)
航空科学基金资助项目(20090557004
20095153022)
西北工业大学校科技创新基金资助项目(2008KJ02022)
关键词
D-S证据理论
最优融合
加权证据距离
全局距离
冲突证据
目标识别
D-S evidence theory
optimal fusion
weighted evidence distance
global distance
conflict evidence
target recognition
