摘要
三维扫描测量机器人在工业现场应用十分广泛,但是受限于机器人较低的定位精度,扫描测量机器人的测量精度不高。为了提高测量精度,采用基于多相机的视觉跟踪系统代替机器人的定位系统,实现扫描测头的精确定位。由于多台大面阵CMOS红外摄像机(2 048 pixel×2 048 pixel)同时采样的数据量大,传统的串行处理器(如PC、DSP等)难以满足数据处理的实时性要求(>30 fps)。以并行运行方式的FPGA作为核心处理单元,采用流水线的方式,实现对三维扫描测头反光标志点中心的定位及匹配。实验结果表明系统数据处理速度满足三维扫描测头位姿跟踪系统的实时性要求。
Three-dimensional scanning measurement robots are widely used in industrial fields.However,scanning measurement robots cannot achieve high-precision measurement,because of the lower positioning accuracy of robots.A multi-camera based visual tracking system was used instead of the robot’s positioning system to achieve accurate positioning of the scanning probe.It is difficult to meet the real-time requirement(>30 fps)with the traditional serial processors,such as PC,DSP,and so on,to process the large amount of data sampled simultaneously by multiple CMOS infrared cameras(2 048 pixel×2 048 pixel).FPGA,with the parallel operation mode,was used as the core processing unit.The positioning and matching of the center of the reflective mark point on the 3 D scanning probe was realized by pipeline.The experiment result shows that the system data processing speed meets the real-time requirement of the 3 D scanning probe pose tracking system.
作者
王文祥
柳燕飞
张瑞
李维诗
WANG Wen xiang;LIU Yan fei;ZHAGN Rui;LI Wei shi(School of Instrument Science and Opto electronic Engineering,Hefei University of Technology,Hefei 230009,China)
出处
《仪表技术与传感器》
CSCD
北大核心
2020年第4期35-39,45,共6页
Instrument Technique and Sensor
基金
佛山市科技创新专项资金项目(2014IT100115)
中央高校基本科研业务费专项资金项目(JZ2018HGBZ0148)。