摩托车轮辐条表面的机械臂自主成像系统设计  

Design of Autonomous Imaging System of Robot Arm on Motorcycle Spoke Surface

作　　者：刘杨 苑玮琦[1] LIU Yang;YUAN Wei-qi(Computer Vision Group,Shenyang University of Technology,Shenyang 110870,China)

机构地区：[1]沈阳工业大学视觉检测技术研究所,辽宁沈阳110870

出　　处：《仪表技术与传感器》2021年第10期90-93,共4页Instrument Technique and Sensor

基　　金：国家自然科学基金资助项目(61271365)。

摘　　要：针对摩托车轮辐条结构多样性导致表面自主成像的空缺问题,设计了一种基于机器视觉的机械臂自主成像系统,能够采集到不同数量、类型的轮辐完整外表面。基于机器视觉的机械臂成像系统分为3部分:首先,机械臂控制相机到初始位置对车轮进行识别和标定;其次,建立成像模型,得到系统硬件的详细参数;最后,机械臂根据计算当前检测车轮的离散成像位姿点进行移位采集图像,并建立图库。经实验证明,该系统控制的机械臂能较好的实现图像的自主采集。Aiming at the vacancy problem of autonomous surface imaging caused by the diversity of motorcycle spokes,a ro⁃botic autonomous imaging system based on machine vision was designed,which can collect different numbers and types of com⁃plete outer surfaces of spokes.The robotic arm imaging system based on machine vision was divided into three parts.Firstly,the ro⁃botic arm controlled the camera to the initial position to identify and calibrate the wheels.Secondly,the imaging model was estab⁃lished to obtain the detailed parameters of the system hardware.Finally,the robotic arm shifted and collected images according to the discrete imaging pose points of the currently detected wheels,and built an image dataset.Experiments proves that the robotic arm controlled by the system can better realize autonomous image collection.

关 键 词：摩托车轮 辐条 机械臂 成像系统 机器视觉 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]