一种改进的星光大气折射观测模型

Modified measurement model of stellar horizon atmospheric refraction

建立精确的观测模型是星光折射间接敏感地平导航法应用的基础和关键。由于大气参数变化的不确定性,国内外现用的星光折射观测模型有一定局限性,使得利用星光折射导航的定位精度大打折扣。根据大气密度随高度、纬度、季节等的变化规律,以及对大气折射原理、平流层大气数据、大气模型的深入研究,建立了连续高度的大气密度模型;在此基础上,改进了现有固定高度单一密度的观测模型,建立了一种自适应连续高度(20～50km)的星光折射观测模型,从整体上提高了自主导航精度及可靠性。同时分析各种摄动对系统状态模型的影响,建立带摄动的系统方程,利用Unscented卡尔曼滤波算法进行计算机仿真研究,并对仿真结果进行误差分析。

The basic and key issue of Stellar Horizon Atmospheric Refraction concept for satellite autonomous navigation is to establish the accurate measurement model. However, the positioning precision of the model at home and abroad is limited due to the uncertainty of the atmospheric parameters. According to the rule that atmospheric density would vary with altitude, latitude, seasons, and by thoroughly studying on the principle of the atmospheric refraction, the data of stratospheric atmosphere, the atmospheric model, the continuous density atmospheric model was set up. On this basis, a self- adapted continuous measurement model （20-50 km） was established through changing fixed altitude and single density, which could improve the precision of the satellite autonomous navigation and the dependability as well in the whole. At the same time, the system equation with the orbit perturbations was established based on analyzing the effect of several sorts of perturbations on the system state model. Then a simulation testing was performed using the Unscented Kalman Filter algorithm, and finally the error source was analyzed.