基于视觉的工业机器人动态分拣研究

发布时间：2018-05-09 14:35

本文选题：动态分拣 + 视觉标定　；参考：《河南科技大学》2017年硕士论文

【摘要】：随着我国生产向智能化转型,智能制造装备的发展逐渐得到重视,工业机器人的应用则是其重要组成部分。由于机器视觉具有非接触性、功能灵活、测量精度高等优点,基于机器视觉的工业机器人系统集成广泛应用于焊接、装配、AGV、测量、分拣等作业中。基于视觉的工业机器人分拣系统研究,尤其是在传送带连续运行下的分拣研究,对智能化生产具有一定的积极意义。在广州数控GR-C型六自由度工业机器人视觉分拣系统的基础上,对硬件进一步改进,通过加装霍尔传感器和工业控制板实现了传送带位置反馈;对视觉标定算法和图像处理算法进行了一定研究:在小孔摄像机理论基础上,结合本系统进行参数简化,设计了一种基于霍夫圆变换和图像区域分割的工业相机标定方法,并通过实验验证了标定精度。在Qt Creator环境下使用Qt框架、OpenCV算法库设计了上位机系统,该系统能够实现相机标定、协调管理视觉系统硬件、实现工业分拣。通过示教编程,在工业机器人上实现了配合上位机和下位机的分拣程序;在MDK集成开发环境下对主控为STM32的工业控制板进行软件编程,实现了基于Free RTOS嵌入式实时系统的分拣系统下位机软件设计;最终对下位机控制策略进行了优化,并通过实验验证了优化效果。通过最终的实验结果,验证了系统的可行性,能够满足在实际生产中对传送带上的物体自动化分拣的需要,对基于视觉的工业机器人动态分拣研究具有一定的理论和应用价值。
[Abstract]:With the transition from production to intelligence in our country, the development of intelligent manufacturing equipment has been paid more and more attention, and the application of industrial robots is an important part of it. Because machine vision has the advantages of non-contact, flexible function and high measuring precision, the industrial robot system integration based on machine vision is widely used in welding, assembling AGV, measuring, sorting and so on. The research of industrial robot sorting system based on vision, especially under the continuous running of conveyor belt, has certain positive significance for intelligent production. Based on the visual sorting system of Guangzhou NC GR-C six-degree-of-freedom industrial robot, the hardware is further improved, and the conveyor belt position feedback is realized by adding Hall sensor and industrial control board. The visual calibration algorithm and image processing algorithm are studied. Based on the theory of small hole camera and the simplified parameters of the system, a calibration method of industrial camera based on Hough circle transform and image region segmentation is designed. The calibration accuracy is verified by experiments. The upper computer system is designed by using QT frame OpenCV algorithm library in QT Creator environment. The system can realize camera calibration, coordinate management vision system hardware, and realize industrial sorting. Through teaching programming, the sorting program with upper computer and lower computer is realized on the industrial robot, and the software programming of industrial control board which is the main control of STM32 is carried out under the environment of MDK integrated development. The software design of sorting system based on Free RTOS embedded real-time system is realized. Finally, the control strategy of lower computer is optimized, and the optimization effect is verified by experiment. Through the final experimental results, the feasibility of the system is verified, which can meet the need of automatic sorting of objects on conveyor belt in actual production, and has certain theoretical and application value for the research of dynamic sorting of industrial robot based on vision.
【学位授予单位】：河南科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41;TP242.2

【参考文献】