An adaptive filtering algorithm based on the quality of information was proposed aiming at the visibility of pulsars in pulsar navigation.According to the analysis of the number of pulsars observed and the quality of information received,the dimension of observation matrix was determined,and an improved Kalman filter was designed to be adaptable to the observation of the information changes.The observed information quality was defined,then filtering equation was constructed according to the definition of pulsar navigation to improve the accuracy and the adaptability of the pulsar.In order to improve the precision of linearization and discretization,Runge Kutta method was used to solve the two body dynamics equation with J2 perturbation.A simulation experiment was performed to verify the improved Kalman filtering algorithm.The experimental results show that after the initial error is corrected,the position precision is 40 m and the velocity precision is 0.019 m/s when one or two pulsars are observed for navigation within 10 000 s.The simulation results show that this filtering algorithm is effective and feasible.