一种证据距离量化及传感器可靠度确定的新方法

A Novel Approach for Calculation of the Evidence Distance and the Evaluation of Sensor Credibility

为解决传统传感器可靠度确定方法中存在的一定条件下证据间距离量化结果不合理的问题,基于信息粒度理论提出了一种新的证据距离——认知距离,并以此提出传感器可靠度的计算方法,用可靠度决定的折扣因子对证据进行折扣,最后使用DS证据理论对修正后的证据进行融合。仿真实验表明,与几种代表性的基于传感器可靠度的融合方法和证据理论及其改进方法对比,本文方法在不同情况下均具有良好性能。

Based on the information granularity theory,the Cognition Distance,a new distance between two bodies of evidence is proposed to solve the problem occurred when using the conventional method which was applied to evaluate the sensor credibility.Besides,the calculation of the credibility is proposed based on it..The evidences are discounted then by the discounting factor which is decided by the evidences＇ credibility.In the end,the modified evidence is fused by using the Evidence theory.Simulation experiments demonstrate the advantages of this new improved algorithm when dealing with different conditions,compared with modified combination rule of Evidence theory and some representative combination rule which based on the sensor credibility.