改进的粒子滤波算法在软着陆小天体中的应用

Application of the Improved Particle Filter Algorithm to Soft Landing on Small Celestial Bodies

针对小天体软着陆过程中估计目标天体质心视线的导航方法失效问题,给出了一种基于特征点信息的自主光学导航方法。为了进一步提高对轨道参数的估计精度,考虑到系统中存在的非高斯噪声,以及基本粒子滤波算法带来的粒子退化和采样枯竭问题,给出了一种改进的粒子滤波算法。最后,通过仿真计算,验证了所给出的自主光学导航方法以及改进的粒子滤波算法的可行性和有效性。

According to the problem of that, the navigation method estimated the line of sight of the centroid of the target celestial body is disabled in the course of soft landing on small celestial bodies, first of all, this paper brings forward a method of autonomous navigation based on the feature point information. At the same time, considering the non-Gaussian noise existing in the system and the problems of the particles degradation or sampling drying up caused by the elementary particle filter algorithm, the improved particle filter algorithm is given to improve the estimate precision of the orbit parameter. Finally, the simulation results indicate that the autonomous navigation method and the improved particle filter algorithm proposed are feasible and efficacious.