In order to overcome lower estimating accuracy and filtering divergence of traditional GMP-PHD algorithm in clutter environment, a new improved GMP-PHD is proposed in this paper, which based on fuzzy hybrid annealed distribution. Compared with the traditional GMP-PHD, the algorithm adopts the state variable decomposition and the introduction of annealing parameters to generate the proposed distribution fuction, and the optimal annealing coefficient is generated by the fuzzy inference system, which could improve the stability and accuracy of particle filter, and then to update a PHD. The simulation show that the improved algorithm can effectively track multiple targets in clutter environment, compared with GMP-PHD filter, closer to the true value, improves the tracking accuracy and system stability.