复杂工业环境的视觉识别与自适应搬运机器人技术研究  

Research on Visual Recognition and Adaptive Handling Robot Technology for Complex Industrial Environment

作　　者：路林艳 黄文涛 张子鑫 Lu Linyan;Huang Wentao;Zhang Zixin(School of Electrical and Control Engineering,Liaoning University of Engineering and Technology,Huludao Liaoning 125000,China)

机构地区：[1]辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛125000

出　　处：《机械管理开发》2024年第12期279-281,共3页Mechanical Management and Development

摘　　要：提出了一种基于视觉识别与自适应搬运机器人技术研究。机器人以TC264为控制器,搭载摄像头,姿态传感器以及驱动模块等,运用目标检测寻找物体,利用卷积神经网络(Convolutional Neural Network,简称CNN)训练工业识别模型,实现摄像头对工业现场物体的识别与分类,再根据麦克纳姆全方位移动优势,进行姿态调整并作用于舵机实现自适应搬运,提高工作效率。A research on a visual recognition and adaptive handling robot technology is proposed.The robot takes TC264 as the controller,equipped with camera,attitude sensor and drive module,etc.It uses target detection to search for objects,uses Convolutional Neural Network(CNN)to train the industrial recognition model,realizes the recognition and classification of the objects on the industrial site by the camera,and then,according to the omni-directional movement advantage of McNamee,carries out the Then according to McNamee's omni-directional movement advantage,it will adjust the attitude and act on the servo to realize adaptive handling and improve the work efficiency.

关 键 词：卷积神经网络 机器视觉 自适应搬运 机器人 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]