Abstract: In order to improve the success rate of robot grasping dynamic workpiece, a dynamic workpiece grasping method of conveyor belt based on machine vision is proposed. Firstly, the system scheme is designed, and the system design goal is expounded.Secondly, the camera debugging, coordinate system calibration and grab point calibration are described and the camera installation target is summarized. The process of coordinate system calibration and grab point calibration is introduced. The grab point is re-calibrated by coordinate system conversion.At the same time, the main interface, TCP communication interface, conveyor belt calibration interface and offset grab point calibration interface are designed, and the pixel resolution calibration interface parameters, the main interface variable parameters of RDE and TPU interaction and Module instruction parameters are explained, which have certain application value.

Key words: Machine vision; Dynamic grasping; Coordinate system conversion; Conveyor belt workpiece

摘要： 为提高机器人对动态工件抓取的成功率，提出了一种基于机器视觉的传送带动态工件 抓取方法。首先对系统方案进行设计，阐述了系统设计目标。其次对相机调试与坐标系标定 以及抓取点标定进行了说明，概述了相机安装目标，介绍了坐标系标定以及抓取点标定的流 程，通过坐标系转换，对抓取点进行重新标定。同时分别对主界面、TCP 通讯界面、传送带 标定界面和偏置抓取点标定界面进行设计，对像素分辨率标定界面参数、RDE 与 TPU 交互的 主要接口变量参数以及 Module 指令参数进行说明，具有一定的应用价值。

关键词: 机器视觉；动态抓取；坐标系转换；传送带工件