高超声速飞行器组合导航鲁棒滤波算法

高超声速飞行器由于其飞行环境的影响，使得GPS和星敏感器的量测噪声表现出非高斯特性。针对常规基于Kalman滤波的组合导航在非高斯噪声下性能下降的问题，提出了基于鲁棒滤波的高超声速飞行器组合导航算法。方法在惯性／GPS／＆异步量测建模的基础上，通过随机去耦将量测更新转化为线性回归问题，并基于M估计获得状态量最优估计。仿真结果表明，方法对非高斯噪声具有更好的鲁棒性，有效提高了高超声速飞行器组合导航系统的性能。

Robust Filter Algorithm of Integrated Navigation for HCV

The measurement noise of GPS and star senror indicate non- Gaussian character, which im- pressed by hypersonic cruise vehicle （HCV）. In order to overcome the perfprmance reduction of Kalman filter under non- Gaussian noise, this paper proposes a kind of robust navigation algorithm for HCV. Base on the INS/GPS/CNS asynchronous measurement model, the algorithm transforms the measerement upda- ting procedure to regression problem by stochastic decoupling. Afterwards, the optimal estimator of INS er- ror statements is obtained on the basis of generalized maximum- likelihood estimator theory, which is ro- bust to the non- Gaussian noise of GPS and star sensor. The simulation results indicate that, the algorithm proposed in this paper has better robust performance than the Kalman filter for the integrated navigation system of HCV.