Abstract: Considering the degeneracy of particle weight and the large amount of calculation existing in the standard particle filtering algorithm, the mean shift algorithm and particle filtering algorithm are fused, then a new particle filtering algorithm is designed based on the mean shift searching algorithm. This approach still obeys the computational outline of the standard particle filtering algorithm. The basic principle of this algorithm is to embed the mean shift searching process into the important sampling process of the particle filtering method via the clustering characteristics of the mean shift algorithm, to have a determinant searching to the particle set, and make each particle converge to local optimal value, approximates the true state distribution by means of the particle clustering of the mean-shift algorithm, and thus achieves good estimation results and improves the status of real time by requiring only a small number of particles compared with the standard PF algorithm on overcoming the defects, such as the degeneracy of the phenomenon of particle weight and the large amount of calculation. The results of a large amount of computational experiments and the GPS / DR integrated navigation experiment show the effectiveness of the new approach.