基于专家系统的弹道导弹目标识别方法

Target Recognition Method of Ballistic Mission Based on Experts Trust

传统Dempster-Shafer(D-S)算法及其改进算法中,一般将冲突证据与不冲突证据统一融合,无论融合过程中冲突证据所占权重多小,这都将降低融合效率。针对这个问题,引入专家系统理论,面向信息融合中证据冲突的问题,将冲突证据与无冲突证据分组,构建了基于专家信任度的弹道导弹目标识别模型和专家知识库模型,从工程上设计了弹道导弹目标识别专家知识库系统,提出了两种专家知识权重确立方法,并通过实例以及仿真,验证了算法的可行性。

In the traditional Dempeter-Shafer algorithm and its improved algorithm, the conflict evidence and the non-conflict evidence are fused together,which will reduce the fusion efficiency regardless of the weight of the conflict evidence in the fusion process. Aiming at this problem, facing the conflict of evidence problem in information fusion, this paper introduces the theory of expert system,divided the evidence into two group,which are conflict and non -conflict evidence,and constructed the ballistic missile target recognition model and expert knowledge base model based on expert trust,and then designed the expert knowledge base system of ballistic missile target recognition based on engineering.Finally,two methods of expert knowledge weight are proposed.Example and simulation showed that this method is effectively.