摘要
针对室内环境下基于ICP匹配的算法比较复杂,运算量大,并且ICP算法高度依赖初始预估,限制了其实际应用范围的问题,提出一种改进的二维激光扫描匹配方法:不需要先验信息,降低特征匹配的复杂度;采用特征线筛选的方法,减少匹配算法运算量。实验结果表明,改进的二维激光扫描匹配方法在保证定位精度不变的条件下,能够提高扫描匹配算法的稳定性和效率,适用于较高精度的室内定位。
Aiming at the problems that the matching algorithm based on ICP is complex with heavy computation,and highly dependent on the initial estimate,which limits its practical application,the paper proposed an improved 2Dlaser scan matching method:the priori information was not necessary so as to reduce the complexity of feature matching;the method of feature line filtering was used to reduce the computation of the matching algorithm.Experimental result showed that the proposed method could improve the stability and efficiency of the scan matching algorithm with the same positioning accuracy,suitable for high-precise indoor positioning.
出处
《导航定位学报》
CSCD
2017年第4期25-29,97,共6页
Journal of Navigation and Positioning
基金
国家重点研发计划项目(2016YFC0803102)
辽宁省高等学校创新团队项目(LT2015013)
关键词
扫描匹配
特征提取
特征匹配
二维激光扫描
室内定位
scan matching
feature extraction
feature matching
2D laser scanning
indoor positioning