Ephemeris Determination of Emergent Spacecraft Using Navigation Data

Abstract

To propagate emergent spacecraft’s flight state, the ephemeris should be determined first. This paper proposes two kinds of algorithms to solve this problem. Primarily, an iteration algorithm based on single navigation data is proposed. Since it is not able to figure out drag coefficient with iteration algorithm, and it is vulnerable to measurement noise, the filter algorithms including Extended Kalman filter and Unscented Kalman filter based on multiple navigation data are established. Both kinds of algorithms are related to SGP4 model, which must use two line elements as inputs to predict position and velocity. Simulations and results show that both algorithms can determine ephemeris effectively and quickly. As for the precision of the iteration algorithm, the propagation deviations are within 1 m for position and 10–3 m/s for velocity within 1 day on three axes when there is no measurement noise. The precision of filter algorithm results is at least 10–5 degree for the four angles, 10–7 for eccentricity, 10–9 Rev/Day for mean motion, and 10–6 for drag coefficient when the measurement noise is 10 m. Use TLEs determined by EKF and UKF to propagate; the position and velocity deviations are within 5 m and 0.005 m/s within 1 day, respectively. Iteration algorithm’s effect becomes poor when measurement noise is added. In contrast, EKF and UKF can eliminate noise’s influence. EKF performs better than UKF under small measurement noise and big initial error condition, whereas under big noise condition, UKF performs better than EKF. Since all the results are convergent to the required precision of TLEs’ format, both approaches can be applied to emergent spacecraft ephemeris determination mission.