Autonomous Navigation Method for Low-Thrust Interplanetary Vehicles

Abstract

An autonomous navigation method for low-thrust interplanetary vehicle is proposed. In the proposed navigation system, one inertial navigation system (INS) is employed to continuously estimate the position and attitude of vehicle, and three X-ray sensors observing X-ray pulsars are utilized to reduce the long-term effects of the errors in the INS. In addition, a modified square-root unscented Kalman filter (MSUKF) is proposed. The MSUKF adopts a fading factor to compensate the impact of modeling error, and contains no linearization steps. The results have shown that the proposed navigation system outperforms the traditional celestial–inertial method and the MSUKF could guarantee a faster convergence compared with former proposed nonlinear filters.