Autonomous navigation of lunar satellite based on the SEM orientation information via EKF

Onboard autonomous navigation can significantly reduce the costs of the mission. This paper deals with the autonomous navigation of a lunar satellite based on the Sun's, the Earth's and the moon's (SEM) orientation information via Extend Kalman Filter (EKF). The initial orbit parameters for EKF are estimated by combining the geometric method with Gauss f-g method. Navigation simulation of the lunar satellite in a circular polar orbit shows that the algorithm provides a position accuracy of better than 1.337 kilometers and velocity accuracy of better than 0.9256 meter per second.