摘要
随着国内外基于工程环境的三维点云处理技术研究的不断深入,开源的PCL(Point Cloud Library)库应运而生并且发展迅速,三维点云配准技术广泛应用在机器视觉、人机交互、无人驾驶等诸多工程领域,且涉及到计算机学、几何计算、传感器等多学科融合。由于传统电铲在手动操作挖掘过程中对于矿场环境感知不足,导致矿山整体开采效率低下,易发生挖掘碰撞、机身倾覆等意外事故。文中基于PCL点云数据处理库,将基于FPFH特征的SAC-IA粗配准,与使用迭代最近点算法(ICP)的精配准方法合并使用,实现矿场环境点云的配准,为后续无人电铲环境感知的研究提供了数据保障。试验结果表明,该方法具有较好的精度与较快的配准速度。
With the continuous in-depth research of 3 D point cloud processing technology based on engineering environment at home and abroad,the open source PCL(Point Cloud Library)library emerged and developed rapidly.3 D point cloud registration technology is widely used in machine vision,human-computer interaction,Unmanned driving and many other engineering fields,and involve the integration of computer science,geometric calculation,sensors and other disciplines.Due to the insufficient perception of the mine environment during the manual excavation of the traditional electric shovel,the overall mining efficiency of the mine is low,and accidents such as excavation collision and overturning of the fuselage are prone to occur.Based on the PCL point cloud data processing library,this paper combines the SAC-IA coarse registration based on FPFH features and the fine registration method using the iterative closest point algorithm(ICP)to realize the registration of the point cloud in the mining environment.The research on environmental perception of unmanned electric shovel provides data guarantee.Experimental results show that this method has better accuracy and faster registration speed.
作者
李光
付涛
张天赐
LI Guang;FU Tao;ZHANG Tian-ci(State Key Laboratory of Mining Equipment and Intelligent Manufacturing,Taiyuan Heavy Industry Co.,Ltd.,Taiyuan 030024;School of Mechanical Engineering,Dalian University of Technology,Dalian 116024)
出处
《机械设计》
CSCD
北大核心
2021年第S01期174-177,共4页
Journal of Machine Design
