室内四旋翼无人飞行器特征提取方法

摘要

针对室内环境下无人四旋翼飞行器同步定位与地图创建时需要进行特征提取的问题，对特征提取的规则进行修改，提出一种新的特征提取方法。该算法具有传统特征提取算法的优点，特征提取速度快、精度高。利用实验室自主研发的无人四旋翼飞行器采集到的激光扫描仪原始数据，进行分割-聚类-线段拟合，得到无人四旋翼飞行器定位、导航需要的特征。实验表明，该算法可以实时、准确、有效地为无人四旋翼飞行器提供定位与地图创建需要的特征。%When four-rotor UAV is locating and mapping simultaneously in indoor environment , it needs to extract environment feature .To response to the problem , this paper puts forward a new method by modifying the rules of feature extraction .The meth-od has the merits of traditional methods , fast calculation speed and high precision .The four-rotor UAV which is designed by the laboratory independently is used to carry out experiment , the raw data which are collected by laser scanner is used to segmenta-tion-clustering and line fitting , then get the feature which is needed to help four-rotor UAV locate and navigate .Results show that the algorithm can accurately and effectively provide feature to UAV for locating and map building .