Research On Autonomous Relative Navigation Algorithm Of Formation Flying Satellite Based On The Dual-quaternion

The technology of formation flying has gained much attention in the aerospace community, as it is a novel approach for space mission to enhance the overall performance compared with the conventional single satellite approach. The formation flying denotes a group of satellites working together in a coordinated manner to accomplish common missions, which therefore demands a fast and accurate determination of relative position and attitude between satellites. The consistency of satellite motions depends on the calculation speed and accuracy, so that a fast and high-accuracy algorithm is desirable. Traditional research on algorithms of formation flying always focuses on the satellite relative orbit, hardly research the relative attitude. Until now, the attitude hasn't been integrated into the orbit, and the relative position and attitude are independently modeling.Comparing with representing the rotation and translation separately, using dual-quaternion to represent the rigid transformation possesses the superiority of concise forms and simple calculation. Given that a rigid body is equivalent to a coordinate system, we propose a unified and original dual-quaternion position and attitude model (P-A model) to describe the satellite's relative position and attitude.In order to obtain more accurate relative position and attitude, we establish a state equation based on P-A model, and then design a GPS-like measurement model to measure the satellite relative position and attitude according to the corresponding measure equation we established. At last, the Extended Kalman Filter(EKF) is applied to eliminate the noise jamming during the state updating and measurement process. All above procedures constitute the autonomous relative navigation algorithm of formation satellites.Considering GPS applied in the formation flying give rise to a couple of problems, we prefer GPS-like. We analysis the origins of these problems and attempts to solve them in GPS-like. Besides, we focus on the contradiction between accuracy and real time, and the fast GPS-like signal-acquisition algorithm is proposed to solve this problem.