China Mechanical Engineering

Previous Articles     Next Articles

Feature Extraction for Map Building of Automated Guided Robot Based on Laser Navigation without Reflector

LI Danyang1,2;ZHANG Ke1,2;XU Ruqing1,2;LUO Zhifeng1,2;WU Jiaqian1,2;GUI Hao1,2   

  1. 1.Shanghai Key Laboratory of Materials Laser Processing and Modification,Shanghai Jiao Tong University,Shanghai,200240
    2.Collaborative Innovation Center for Advanced Ship and Deep-Sea Exploration,Shanghai,200240
  • Online:2018-11-25 Published:2018-11-27

[供应链调度]无反射板激光导航机器人地图创建的特征提取

李丹阳1,2;张轲1,2;许如清1,2;罗志峰1,2;吴佳谦1,2;桂豪1,2   

  1. 1.上海交通大学上海市激光制造与材料改性重点实验室,上海,200240
    2.高新船舶与深海开发装备协同创新中心,上海,200240

Abstract: To realize the line feature extraction of map building for laser navigation robot without reflector,a stepwise decomposition method was proposed herein.The line feature extraction method was divided into 3 steps: breakpoint detection,line segmentation,and line extraction.Firstly,an adaptive threshold method was used to detect the breakpoints.And then,a point set was separated,and the line segment was segmented based on the iterative adaptive point algorithm.Finally,a least square method was used to fit the lines,and the region search method was used to optimize the line feature extraction.Experiments show that the repetitive positioning accuracy of the algorithm is within ±6mm and the feature extraction time is not longer than 0.02s,which meets the actual navigation requirements of robots.

Key words: robot, feature extraction, laser navigation, non-reflector, map building

摘要: 针对无反射板激光导航机器人地图创建的直线特征提取,采用逐步分解的方法,将直线特征提取分为断点检测、线段分割、直线提取三个步骤逐步分离点集。首先采用自适应阈值法进行断点检测;然后基于迭代适应点算法进行点集分离、线段分割;最后采用最小二乘法拟合直线并结合区域搜索法进行优化,进一步提高直线特征提取的精度。实验表明,算法的重复定位精度在±6 mm以内,特征提取时间不大于0.02 s,满足机器人的实际导航需求。

关键词: 机器人, 特征提取, 激光导航, 无反射板, 地图构建

CLC Number: