摘要
针对月球所具有的特殊环境,将两个航天器利用相对位置信息导航的方法应用到环月航天器的自主导航中。由于低轨道环月的摄动力模型受复杂性因素的影响,动力学模型存在较大的误差,导致滤波发散,就此提出一种自适应衰减因子非线性滤波方法。通过仿真结果比较表明,此方法不仅能更好地消除非平稳随机噪声带来的长期项状态估计误差,还能适应采样周期的实时变化进而保证最佳的衰减抑制效果。
An autonomous navigation system based on the measurements of relative position vectors from one spacecraft to another is applied to determine the orbital states of the two lunar probes, Due to the errors in the model of dynamics and process noises, more realistic and complex orbital perturbation factors caused a divergent results are considered. To solve this problem, a modified unscented Kalman filter based on the fixed and adaptively-estimated fading memory factors is proposed. The compared simulation results demonstrate the latter method is not only good at eliminating the longtime estimated errors, but also ensuring the optimal filtering results when the sample interval varies with the time.
出处
《航天控制》
CSCD
北大核心
2009年第3期23-27,共5页
Aerospace Control
关键词
自主导航
环月航天器
UKF滤波
衰减记忆
Autonomous navigation
Lunar probe
Unscented Kalman filter
Fading memory
