基于点云的空间非合作目标运动测量技术研究  

Motion Measurement Technology Based on Point Cloud for Space Non-cooperative Target

作　　者：高伟[1] 杨光[1] 

机构地区：[1]北京航空航天大学机械工程及自动化学院,北京100191

出　　处：《计算机仿真》2016年第6期41-45,58,共6页Computer Simulation

基　　金：国家"973"重点基础研究发展项目(2013CB035503)

摘　　要：在空间非合作目标准确探测和捕获问题的研究中,存在目标模型完全未知以及运动情况不明的问题。根据空间非合作目标的上述特点,提出了采用点云的非合作目标从模型获取到相对运动测量的技术方法。首先提出一种曲率度量提取关键点并结合特征提取的方法完成多视点云配准,之后针对运动测量提出了一种哈希索引表方法。利用多视点云配准获取的整体点云模型建立哈希索引表,待求姿态的点云在哈希索引表中查询匹配,获得目标的相对姿态。最后软件系统仿真结果表明,多视点云配准结合哈希索引表的方法,速度快,精度高,满足目标对准测量实时性和精度的要求。Although non - cooperative space target detection and capture have important application prospects, there are big difficulties that target model is completely unknown and the motion is unknown. Based on this character- istic of non - cooperative target space, propose a complete technical method for measuring relative motion of non - co- operative target, from obtain point cloud to relative motion measurement. First proposed a measure based on the cur- vature of the key points extracted and combined feature extraction method to complete multi - view point cloud regis- tration, and motion measurement section presents a hash index table methods. Using multi - view point cloud model point cloud establish quasi - acquired hash index table, then query matches from the hash table to estimated point cloud so that calculate the motion condition. Finally, the simulation results show that the software system, combining multi - view point cloud registration method based on the hash index table, high speed, high precision, meet the re- quirements of real - time engineering and precision.

关 键 词：空间非合作目标 运动测量 点云配准 哈希索引表 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP391.7[自动化与计算机技术—计算机科学与技术]