中文摘要：

伪卫星作为卫星导航系统的重要增强手段，能够克服卫星系统在导航定位及武器精确制导方面存在的缺点。针对伪卫星空中基站位置不易精确确定的问题，设计了临近空间伪卫星空中基站定位SINS／CNS／SAR组合导航系统，建立了组合导航非线性数学模型；在吸收抗差自适应滤波、高斯滤波和粒子滤波优点的基础上，提出了一种抗差自适应高斯混合Sigma点粒子滤波算法。将提出的算法应用于SINS／CNS／SAR自主导航系统进行计算仿真，并与Unscented卡尔曼滤波和粒子滤波比较。结果表明：提出的新算法能够满足伪卫星空中基站自主导航定位的需求，导航精度明显高于Unscented卡尔曼滤波和粒子滤波算法。

英文摘要：

Pesudolite, as an important means to enhance the satellite navigation system, can overcome the disad- vantage of satellite navigation system in terms of positioning and precision-guided weapons. For pseudolites air based positioning that is not easy to accurately determine the location, we design a SINS/CNS/SAR integrated navigation system for the air based positioning of near space pseudolite, and the non-linear mathematical model of the system is established. On the basis of absorbing the merits of robust adaptive filtering, Gaussian filtering and particle filte- ring, a robust adaptive Gaussian mixture sigma point particle filter algorithm is proposed. Then, the proposed algo- rithm is applied to SINS/CNS/SAR autonomous navigation system for calculation and simulation and the results are compared with those of unscented Kalman filtering and particle filtering. The results and their analysis show prelimi- narily that the new algorithm can meet the and the navigation accuracy is significantly algorithms. needs of autonomous navigation positioning of the pseudolite air based higher than those of the unscented Kalman filtering and particle filtering algorithms.