Abstract: A high accuracy integrated navigation method under long\|endurance environment is researched by using strapdown inertial navigation system (SINS), star sensor and Bei Dou receiver in this paper. Errors of SINS, misalignments of star sensor and positioning errors of Bei Dou are modeled. Differences between attitudes output by star sensor and by SINS and differences between positions output by Bei Dou receiver and by SINS are chosen as measurements, and the measurement equation of integrated navigation is constructed. To solve the problem that statistic characteristics of measurement noise is probably uncertain under long\|endurance environment, a simplified Sage\|Husa self\|adaptive filtering algorithm is adopted in the design of integrated navigation filter. Simulation results show that, position accuracy of this integrated navigation method reaches ±11.86m, attitude accuracy reaches ±0.27′.

Key words: Strapdown inertial navigation system, Star sensor, Bei dou receiver, InteGrated navigation, Self-adaptive filter