基于激光雷达的移动机器人实时位姿估计方法研究  

Research on Laser Radar Based Real-Time Pose Estimation for Mobile Robots

作　　者：杨明[1] 董斌[1] 王宏[1] 张钹[1] Helder Araújo 

机构地区：[1]清华大学智能技术与系统国家重点实验室,北京100084 [2]葡萄牙Coimbra大学系统与机器人研究所

出　　处：《自动化学报》2004年第5期679-687,共9页Acta Automatica Sinica

基　　金：国家863计划(2002AA420110);科技部中葡科技合作计划(2001066)资助~~

摘　　要：位姿估计是移动机器人研究中的一个关键问题,对于运动目标跟踪、机器人导航、地图生成等具有重要意义.提出了一种基于二维激光雷达的实时位姿估计方法,包括基于Hough变换的切线角度直方图算法和迭代切线加权最近点算法两部分.两种算法的结合和切线信息的引入,有效地解决了传统方法无法解决的局部最小值、类孔径、大计算量等问题.仿真数据和室外环境实际数据的大量实验结果表明它具有精度高、速度快、适用性广和对噪声、遮挡和类孔径问题鲁棒性高的特点.Pose estimation is a key issue in mobile robots and important to moving objects tracking, robot navigation, map building, etc. A real-time pose estimation method using 2D range data from a laser radar is presented, which consists of iterative tangent weighted closest point and Hough transform based tangent angle histogram algorithms to overcome problems with past methods, such as local minimum, aperture-like, high computation problem, etc. This method has been tested on both synthetic and real range data collected in outdoor environment. Experimental results demonstrate its high accuracy, low computation, wide applicability and high robustness to aperture-like problem, occlusion and noises.

关 键 词：位姿估计 角度直方图 迭代最近点 切线信息 移动机器人 激光雷达 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]