Abstract:

The evidence theory is widely applied in the information fusion system due to its flexibility in representing and handling uncertainty information. However, the classical evidence theory assumes that evidence is independent, which is unrealistic. The ignorance of dependence among evidence may cause an unreasonable result. Thus, how to combine dependent evidence effectively is an inevitable and important issue. In this paper, the combination methods of dependent evidence are divided into two categories from the points of the modification of the Dempster’s combination rule and the modification of evidence source. The basic ideas, advantages and disadvantages of these methods, as well as their relationships are analyzed. Finally, the shortcomings of the present methods and the possible research directions in the future are discussed.