A method for combining conflicting evidences with improved distance function and Tsallis entropy
Hanwen Li, Fuyuan Xiao
Abstract
For the sake of great ability of handling uncertain information, Dempster-Shafer evidence theory is extensively used in information fusion. Nevertheless, when there exists highly inconsistent evidences, using classical Dempster's combination rule may lead to counter-intuitive results. To address this issue, a new conflicting evidences combination method based on distance function and Tsallis entropy is proposed. Numerical examples are used to illustrate the feasibility and efficiency of the proposed method. Further, an fault diagnosis problem is used as an example to show the effectiveness and superiority of the proposed method. The proposed method outperforms other methods that the proposed method recognize the target by the probability 99.49%, which is higher than other methods.