In order to avoid the sample impoverishment and degradation phenomenon of the traditional particle filter algorithm, the particle filter algorithm based on the gravitation field(GFA) is proposed to optimize the resampling process of the particle filter. In the gravitation field algorithm particle filter(GFA-PF), a mobile factor which can drive the particles approach to high likelihood region is proposed, so that the particles can concentrate in the real state nearby rapidly. Meanwhile, the proposed rotation factor makes the particles around the true state keep certain distance randomly, which can avoid excessive concentration, so that the diversity of particles is increased. The simulation results show that the proposed algorithm is effective, and it has fast convergence speed, high estimation accuracy and strong robust performance.