摘要
针对室内移动机器人的定位问题,利用二维扫描激光获得周边环境的点云数据,分别采用Trimmed ICP方法、基于角点特征匹配的HAYAI方法以及基于点-直线特征匹配的PLICP方法,并以此三种不同的扫描匹配算法求解连续两次测量帧之间的相对位移估计,以此对里程计定位结果进行有效校正,从而可提高室内移动机器人长距离定位的精度.仿真对所采用的方法从实时性和精度两个方面进行了比较,结果表明了各方法的有效性和适用范围.
The localization of indoor mobile robots was studied.Point cloud data of surrounding environments were obtained by a two-dimensional laser range finder,and then three scan-matching algorithms,including Trimmed ICP(iterative closest point)algorithm,corner feature matching based HAYAI algorithm,as well as point-line matching based PLICP algorithm,were used respectively to estimate the relative displacement between two consecutive measurements to calibrate the localization result from odometer approaches.Better accuracy of long-range localization of indoor mobile robot could be achieved thereby.In the simulation,both the real-time property and accuracy of the mentioned three methods used were compared,showing the effectiveness and application field of the methods.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2013年第S1期262-266,共5页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(61005092
61105094)
高等学校博士学科点科研基金资助项目(20100092120026)
复杂工程系统测量与控制教育部重点实验室开放课题(MCCSE2012B04)
关键词
移动机器人
激光测距仪
扫描匹配
定位
迭代最近点
mobile robot
laser range finder
scan matching
localization
iterative closest point
