Autonomous orbit determination of navigation constellation based on cross-link range is studied in this paper. Considering the characteristics of near-circular orbits of navigation satellites and onboard computing ability, a set of nonsingular orbital elements are selected as state variables. Since the state variables are slow variables on the whole, maximum integration step used in the proposed algorithm, it can expand to 60 minutes, which is much longer than that of other methods. Meanwhile, instead of solving high-order variable coefficient linear differential equations using numerical method, the state transition matrix can directly refer to the result of analytical method. These two improved aspects can reduce the calculation burden of onboard computers obviously. And then the improved extended Kalman filter (EKF) is utilized to estimate the satellite position and velocity vectors of the constellation by fusing cross-link range data and satellite orbital dynamic information. Simulation results show that the proposed algorithm is feasible to improve the calculation efficiency while the orbit determination accuracy is satisfied, which is very important for realtime onboard data processing.
展开▼
