Abstract: This paper focuses on the high accuracy navigation method based on the information fusion of star camera and inertial measurement unit (IMU) for aerial vehicles. Among the existing navigation methods for aerial vehicles, the integrated navigation of inertial navigation system (INS) and global navigation satellite system (GNSS) suffers from the risk of performance degradation in the radio signal denied environment. In the traditional INS/CNS (celestial navigation system) integrated navigation system, the drift of the gyroscopes in the inertial measurement unit can be suppressed, while the zero bias of the accelerometers cannot be eliminated effectively. In order to cope with this problem, a novel autonomous navigation approach based on the line of sight (LOS) measurements of space targets is presented, where the LOS vectors of space targets and background stars with known ephemeris are observed by the star camera, and the state of the vehicle is predicted by using the IMU. The position, velocity and attitude of the vehicle are estimated together with the calibration of the IMU measurement bias via the extended Kalman filter (EKF). In addition, an optimal selection strategy of observed targets based on the Cramer Rao lower bound (CRLB) is designed. The effectiveness of the presented method is illustrated through the visibility analysis of the space targets, the observability analysis of the navigation and the numerical simulation of the navigation filter.

Key words: aerial vehicle, navigation, space target, optimal selection, Kalman filter