According to the systematic error of sensors data, a new data fusion algorithm is proposed based on the evidence
theory. The proposed algorithm divides sensor data into groups according to the deviation, and the concept of basic
probability assignment is applied to generate the frame of discernment. The combination rule is also applied, and the Mass
function of the combined evidence is referred as the weight assignment function of all data. Finally, the fusion result is
obtained by weighted summation. Simulation results show the effectiveness of the proposed algorithm.