基于人工路标的AGV识别与定位技术研究  

Research on AGV Recognition and Positioning Technology Based on Artificial Road Signs

作　　者：陈成 钱东海[1] 赵伟[1] 刘洋[1] 孙林林 CHEN Cheng;QIAN Dong-hai;ZHAO Wei;LIU Yang;SUN Lin-lin(School of Mechatronic Engineering and Automation,Shanghai University,Shanghai 200444,China)

机构地区：[1]上海大学机电工程与自动化学院,上海200444

出　　处：《自动化与仪表》2020年第10期73-77,91,共6页Automation & Instrumentation

摘　　要：针对结构化的工厂环境中AGV到达目标点附近时需要在调整位姿后准确停靠的问题,设计了专门用于定位的新型视觉人工路标,提出一种稳定的路标检测算法。利用路标外围的红色特征进行阈值分割得到二值化图像;采用形态学运算、边缘检测对图像进行处理;利用轮廓检测获得候选区域并作裁剪处理;对笛卡尔坐标系下的图像进行对数极坐标变换;通过计算直方图距离识别路标,确定路标4个顶点的图像坐标与世界坐标的关系,利用四对匹配点求解出AGV的位姿。试验结果表明,路标检测具有很好的鲁棒性,定位精度满足AGV定位需要。Aiming at the problem that AGV needs to adjust the posture and accurately stop after reaching the target point in structured factory environment,a new type of visual artificial road sign specially designed for positioning is designed,and a stable road sign detection algorithm is proposed.The red features around the road sign are used to perform threshold segmentation to obtain a binarized image;use morphological operations and edge detection to process images;use contour detection to obtain candidate regions and cut them out;performs log-polar transformation on the image in Cartesian coordinate system;recognizes road signs by calculating the histogram distance,and the relationship between the image coordinates of the four vertices of the road sign and the world coordinates is determined,and the four pairs of matching points are used to solve the AGV pose.The experimental results show that the road sign detection has good robustness and the positioning accuracy meets AGV positioning needs.

关 键 词：视觉定位 自动导引小车 目标识别 人工路标 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]