上甑机器人构型及其多点图像采集系统研究

Research on the steamer-filling robot configuration and multi-point image acquisition system

针对现有的自动上甑设备存在的机器人成本高、工作空间局限、对图像采集设备性能要求高等问题,设计了一款五自由度专用上甑机器人,实现了多工位之间的切换作业,摆头末端的三维准确定位及摆动高度差的补偿等功能,并利用MATLAB仿真验证了其工作空间。热像仪安装于可回转的铺料头上,采用多点图像采集再拼接的方法实现了甑桶料层表面温度信息采集;通过图像区域分块、阈值化处理和特征点匹配对两级筛选的方法对基于SURF算法的图像拼接方法进行优化改进,提高了图像拼接的实时性和准确性,最后用实验证明了改进方法的有效性。

Aiming at the high cost of the robot itself,limited workspace and high performance requirements for image acquisition equipment in automatic steamer-filling equipment,it designs a five degree of freedom steamer-filling robot,realizes the functions of switching between multiple operation stations,3D accurate positioning and compensate for the height difference caused by wobble.Based on MATLAB it simulates and verifies the workspace.The thermal imager is mounted on a revolving paving mechanism,multi-point image acquisition and mosaic realize the collection of surface temperature information of barrel.It takes the image region partitioning,threshold processing and feature matching pair two-stage screening to optimize the image mosaic method based on SURF algorithm.The experiments prove the effectiveness of the improved method,improve the real-time accuracy of image mosaic.