矿井下三维场景激光扫描系统研究与实现

发布时间：2018-04-25 06:27

本文选题：智能开采 + 三维激光扫描　；参考：《华中科技大学》2015年硕士论文

【摘要】：我国煤炭资源储量逐年下降,开采技术落后导致的资源浪费现象却日益严重,因此,煤炭智能化高效开采技术的研究越来越受到重视。针对煤炭资源短缺、开采效率不高以及自动化智能化程度较低等问题,本文将三维激光扫描测量技术引入到煤炭智能开采中,在分析了系统扫描测量原理与采掘现场空间信息的采集、处理和传输等环节的基础上,研发出一套应用于煤矿井下场景测量的三维激光扫描系统,为实现智能开采提供了完善可靠的采场环境信息支持。系统采用高速激光测距传感器AR2500,配合高精度角度编码器,在步进电机驱动下构成三维激光扫描结构。考虑到系统的实时性要求和应用前景,本文设计了一种基于ARM Cortex-M4微控制器的嵌入式点云采集控制系统,并移植了小型开源协议栈Lw IP(Light Weight IP)来实现与上位机软件的数据传输和指令传送。上位机软件借助开源应用框架Qt编写,通过标准网络协议与扫描装置通信,实现点云接收、控制命令发送、点云显示、点云预处理和数据存储等功能。系统在获取点云的过程中,由于器件测量误差和环境干扰等因素,生成的点云叠加了大量随机噪声。为了获得高质量的点云,深入研究了点云的降噪、平滑和简化等预处理技术,提出了高效的预处理算法,取得了理想的效果。此外,针对系统固有的测量误差,在分析了误差来源的基础上,提出了系统参数的修正与校准实验思路,从软件算法层面提高了系统的测量精度。通过本文设计实现的三维激光扫描系统,可以快速准确地获取矿井下的三维场景可视化模型。体积小、采样速率高、成本低廉和方便联网等特点,使得该系统具有广阔的应用前景和研究价值。
[Abstract]:The reserves of coal resources in China are decreasing year by year, but the phenomenon of resource waste caused by backward mining technology is becoming more and more serious. Therefore, more and more attention has been paid to the research of intelligent and efficient coal mining technology. In view of the shortage of coal resources, the low efficiency of mining and the low degree of automation and intelligence, this paper introduces the 3D laser scanning measurement technology into the intelligent mining of coal. Based on the analysis of the principle of system scanning measurement and the acquisition, processing and transmission of spatial information in mining field, a 3D laser scanning system is developed for underground coal mine scene measurement. It provides perfect and reliable environmental information support for intelligent mining. The system uses high speed laser ranging sensor AR2500 and high precision angle encoder to construct a three dimensional laser scanning structure driven by step motor. Considering the real-time requirement and application prospect of the system, an embedded point cloud acquisition and control system based on ARM Cortex-M4 microcontroller is designed in this paper. And the small open source protocol stack LW IP(Light Weight is transplanted to realize the data transmission and instruction transfer with the host computer software. The upper computer software is written with the open source application framework QT and communicates with the scanner through standard network protocol to realize the functions of point cloud receiving, control command sending, point cloud display, point cloud preprocessing and data storage. In the process of obtaining point cloud, the generated point cloud is superimposed with a large amount of random noise due to the measurement error of the device and the environmental interference. In order to obtain high quality point cloud, the preprocessing techniques such as noise reduction, smoothing and simplification of point cloud are deeply studied, and an efficient preprocessing algorithm is proposed, and the ideal effect is obtained. In addition, according to the inherent measurement error of the system, based on the analysis of the error source, the experimental idea of system parameter correction and calibration is put forward, which improves the measurement accuracy of the system from the level of software algorithm. The 3D laser scanning system designed in this paper can quickly and accurately obtain the 3D scene visualization model under the mine. Because of its small size, high sampling rate, low cost and convenient networking, the system has broad application prospect and research value.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD82;TN249

【参考文献】