Aiming at the co-navigation problem of master-slave UAV formation, the motion model of master-slave UAV formation is established, and the flicker noise is added to the observation model. The particle filter algorithm is used to simulate the experiment, which makes the experimental result more realistic. Aiming at the phenomenon of particle depletion in particle re-sampling process, the weight influence factor is introduced in the process of particle weight normalization, and the theoretical proof is given, and compared with the standard particle filter algorithm. The simulation results show that under the flicker noise, the improved particle filter makes the navigation precision of the master-slave UAV formation significantly improved, which has certain practical value.