采用混合粒子群算法的星图识别方法

Star recognition method based on hybrid particle swarm optimization algorithm

为提高大视场高灵敏度星敏感器的星图识别速度和识别成功率,提出了一种基于混合粒子群算法的星图识别方法,该方法首先根据星图中星点的灰度信息确定候选识别主星集合;然后选择该集合中的一个星点为圆心,以一定角距为半径画圆,将圆内的所有星点构成特征数据集合;然后利用混合粒子群算法对圆内的星点进行快速路径寻优;最后利用最优路径长度进行索引,并利用最优路径中前三个星点间的角距以及它们的星等信息进行匹配识别;实验结果表明,与现有识别方法相比,该方法具有高的识别率,良好的实时性和鲁棒性,且所需的导航星库容量小。

A new star recognition method based on hybrid particle swarm algorithm was developed to increase recognition speed and success rate for large field high sensitivity star sensors. Firstly, several candidate recognition main stars were determined with the gray information. Then a circle was drawn with the given circle radius, and all stars in the circle were selected to constitute a characteristics data collection. Hybrid particle swarm algorithm was used for fast path optimization to construct recognition characteristics. Finally, the optimal path length was used for indexing to search matching star, and preceding three star angular distance and magnitude in the optimal path were used for matching recognition to enhance recognition speed and success rate. Experimental results show that, compared with existing recognition methods, star recognition method based on hybrid particle swarm optimization algorithm has a higher recognition rate, good real﹣time and robustness to noise, and it requires small star database capacity.