论文部分内容阅读
针对现有基于D-S融合信息进行决策存在的问题,提出一种基于不确定融合证据的决策方法。该方法结合基元属性和非基元属性进行决策。首先给出了对候选决策集进行精简的原则,建立了焦元属性对目标决策属性的属性层面支持度以及证据层面支持度计算模型;然后对决策属性的决策向量、决策矩阵、理想状态向量进行定义,给出决策属性总体支持度的计算方法;最后对多源目标识别融合结果进行决策分析。结果证明了该方法的合理性与有效性,它同时具有对高冲突证据融合结果与低冲突证据融合结果的决策能力。
Aiming at the existing problems of decision making based on D-S fusion information, a decision-making method based on evidence of uncertain fusion is proposed. The method combines primitive attributes and non-primitive attributes for decision making. Firstly, the principle of streamlining the candidate decision set is given, and the attribute level support of the focal element attribute to the target decision attribute and the support level calculation model of the evidence level are established. Then the decision vector, decision matrix and ideal state vector of the decision attribute are established Definition, and gives the calculation method of the overall support degree of decision attributes. Finally, the multi-source target recognition fusion results are analyzed. The result proves the rationality and validity of this method. It also has the ability of decision-making on the result of fusion of evidence with high conflict and evidence with low conflict.