基于三级多源信息融合结构的SINS性能测试综合评估算法  

Comprehensive evaluation algorithm of SINS performance testing based on three-level multi-source information fusion structure

作　　者：李军伟[1,2] 程咏梅[1] 陈克喆[1] 

机构地区：[1]西北工业大学自动化学院,西安710072 [2]河南大学计算机与信息工程学院,开封475004

出　　处：《中国惯性技术学报》2014年第2期177-184,共8页Journal of Chinese Inertial Technology

基　　金：航空科学基金项目(20100853010)

摘　　要：针对SINS性能测试存在指标多、不确定性等问题,提出了一种基于三级多源信息融合结构的SINS性能测试综合评估算法。给出了包括精度测试、稳定性测试和可靠性测试等三大性能测试指标和16个性能测试底层子指标构成的SINS性能测试指标体系,并构建SINS测试评估结构框架。利用多个专家对底层性能测试子指标进行评估给出相应性能子指标的模糊评价,并构造基本置信指派组成多个证据;利用基于局部冲突分配的组合规则进行三级证据融合,得到性能测试指标的综合评价结果。仿真实验结果表明:与传统的证据理论融合方法相比,该算法可以较为全面地很好地评估SINS性能,而且结果更为合理。In view of the problems in SINS performance test, such as uncertainty and too many indices etc., a comprehensive evaluation algorithm of the SINS performance test is proposed based on three-level multi-source information fusion structure. The proposed index system of the SINS performance test includes accuracy, stability and reliability performance testing index and 16 performance testing bottom subindices. Many experts are invited to evaluate the credibility of the bottom subindex data in order to obtain the fuzzy assessment corresponding to subindex and construct the mass function of each expert's evidence. The structural framework of the SINS performance testing and evaluation is built. The improved combination rule based on local conflict distribution can be used to obtain the results of SINS performance comprehensive testing and evaluation by three-level evidence fusion. The simulation results show that, compared with the traditional Dempster-Shafer theory (DST) method, the proposed algorithm can not only evaluate the SINS performance, but also enhance the reliability of the evidence combination results.

关 键 词：捷联惯性导航系统 性能评估 专家系统 模糊评判 信息融合 

分 类 号：U666.1[交通运输工程—船舶及航道工程]