基于视觉伺服的三轴机械装置控制
发布时间：2018-12-06 07:45
【摘要】：自动化技术在制造业中的广泛应用,提高了产品质量,改善了劳动条件,减少了原料和能源的浪费,三轴机械装置在其中发挥了重要作用。而传统三轴机械装置只能根据规划的运动模式重复性的工作,这对于现在小批量多品种的加工方式变得尤为不便。为了使三轴机械装置对不同环境有更高的适应性,近些年基于视觉伺服的运动控制成为了研究热点。本文主要针对电子消费产品装配线的视觉检测,分析目前市场上三轴机械装置存在的不足,设计了基于视觉伺服的三轴机械装置运动控制方案。本文的主要内容包含三个模块:三轴机械装置运动控制方案总体设计、视觉定位和运动控制。在三轴机械装置运动控制方案总体设计模块,三轴机械装置采用龙门架式的框架设计方案,使用运动控制卡和电机驱动器结合的运动控制方式。同时使用一台摄像机位于机械装置顶部,用于待检测目标的定位,另一台摄像机安装在Z轴上,用于视觉扫描和视觉反馈。通过两台摄像机配合,实现对待检测目标的精确定位和扫描识别。在视觉定位模块,为了根据图像信息得到待检测目标在运动机构坐标系中的坐标。首先对视觉系统标定,获取图像坐标系到运动机构坐标系的变换关系。在获取待检测目标在图像中的坐标时,首先使用ORB(oriented FAST and rotated BRIEF)匹配算法在全局图像中定位模板图像的坐标和旋转角度,然后根据示教获取的模板图像与待检测点位之间的相对坐标计算出各点位在全局图像中的坐标。在运动控制模块,本文首先介绍了运动控制卡和电机驱动器的参数配置,然后介绍了电机速度设置和运动脉冲计算。然后介绍了基于爬山法的自动对焦技术和基于遗传算法的路径规划。最后,根据待检测目标在扫描相机视野中的坐标作为视觉反馈,调整运动机构,确保待检测目标在扫描相机视野中。由于需要获取模板图像及其与待检测点位之间的相对坐标等先验信息,本文还设计了示教软件。实验结果表明,本文设计的基于视觉伺服的三轴机械装置,能实现对不同高度、多点位待检测目标的精确定位。
[Abstract]:The wide application of automation technology in manufacturing improves the quality of products, improves the working conditions and reduces the waste of raw materials and energy, in which triaxial machinery plays an important role. But the traditional three-axis mechanical device can only work repeatedly according to the planned motion mode, which is especially inconvenient for the small batch and many kinds of processing methods. In order to improve the adaptability of triaxial mechanical devices to different environments, the motion control based on visual servo has become a research hotspot in recent years. This paper mainly aims at the visual inspection of the assembly line of electronic consumer products, analyzes the shortcomings of the three-axis mechanical device in the market at present, and designs a motion control scheme of the three-axis mechanical device based on visual servo. The main contents of this paper include three modules: the overall design of motion control scheme of three-axis mechanical device, visual positioning and motion control. In the overall design module of three-axis mechanical device motion control scheme, the three-axis mechanical device adopts the frame design scheme of gantry frame, and the motion control mode which combines motion control card and motor driver. At the same time, one camera is located at the top of the mechanical device, which is used to locate the target to be detected, and the other camera is mounted on the Z axis for visual scanning and visual feedback. Through the cooperation of two cameras, the accurate location and scanning recognition of the detection target are realized. In the visual positioning module, in order to get the coordinates of the target to be detected in the moving mechanism coordinate system according to the image information. First, the vision system is calibrated to obtain the transformation relationship between the image coordinate system and the kinematic mechanism coordinate system. In order to obtain the coordinates of the target to be detected in the image, the ORB (oriented FAST and rotated BRIEF) matching algorithm is used to locate the coordinates and rotation angles of the template image in the global image. Then the coordinates of each point in the global image are calculated according to the relative coordinates between the template images and the points to be detected. In the motion control module, this paper first introduces the parameter configuration of the motion control card and motor driver, then introduces the speed setting of the motor and the calculation of the motion pulse. Then the automatic focusing technology based on mountain climbing and the path planning based on genetic algorithm are introduced. Finally, according to the coordinates of the target to be detected in the scan camera field as visual feedback, the motion mechanism is adjusted to ensure that the target to be detected in the scan camera field of vision. Due to the need to obtain the prior information such as the template image and the relative coordinates between the template image and the point to be detected, the teaching software is also designed in this paper. The experimental results show that the three-axis mechanical device designed in this paper based on visual servo can accurately locate the target with different heights and multiple points to be detected.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH16;TP391.41
本文编号：2365691
[Abstract]:The wide application of automation technology in manufacturing improves the quality of products, improves the working conditions and reduces the waste of raw materials and energy, in which triaxial machinery plays an important role. But the traditional three-axis mechanical device can only work repeatedly according to the planned motion mode, which is especially inconvenient for the small batch and many kinds of processing methods. In order to improve the adaptability of triaxial mechanical devices to different environments, the motion control based on visual servo has become a research hotspot in recent years. This paper mainly aims at the visual inspection of the assembly line of electronic consumer products, analyzes the shortcomings of the three-axis mechanical device in the market at present, and designs a motion control scheme of the three-axis mechanical device based on visual servo. The main contents of this paper include three modules: the overall design of motion control scheme of three-axis mechanical device, visual positioning and motion control. In the overall design module of three-axis mechanical device motion control scheme, the three-axis mechanical device adopts the frame design scheme of gantry frame, and the motion control mode which combines motion control card and motor driver. At the same time, one camera is located at the top of the mechanical device, which is used to locate the target to be detected, and the other camera is mounted on the Z axis for visual scanning and visual feedback. Through the cooperation of two cameras, the accurate location and scanning recognition of the detection target are realized. In the visual positioning module, in order to get the coordinates of the target to be detected in the moving mechanism coordinate system according to the image information. First, the vision system is calibrated to obtain the transformation relationship between the image coordinate system and the kinematic mechanism coordinate system. In order to obtain the coordinates of the target to be detected in the image, the ORB (oriented FAST and rotated BRIEF) matching algorithm is used to locate the coordinates and rotation angles of the template image in the global image. Then the coordinates of each point in the global image are calculated according to the relative coordinates between the template images and the points to be detected. In the motion control module, this paper first introduces the parameter configuration of the motion control card and motor driver, then introduces the speed setting of the motor and the calculation of the motion pulse. Then the automatic focusing technology based on mountain climbing and the path planning based on genetic algorithm are introduced. Finally, according to the coordinates of the target to be detected in the scan camera field as visual feedback, the motion mechanism is adjusted to ensure that the target to be detected in the scan camera field of vision. Due to the need to obtain the prior information such as the template image and the relative coordinates between the template image and the point to be detected, the teaching software is also designed in this paper. The experimental results show that the three-axis mechanical device designed in this paper based on visual servo can accurately locate the target with different heights and multiple points to be detected.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH16;TP391.41
【参考文献】
相关期刊论文 前3条
1 赵辉,鲍歌堂,陶卫;图像测量中自动调焦函数的实验研究与分析[J];光学精密工程;2004年05期
2 李奇,冯华君,徐之海,边美娟,申溯,戴瑞春;数字图象清晰度评价函数研究[J];光子学报;2002年06期
3 赵东波,熊有伦;机器人离线编程系统的研究[J];机器人;1997年04期
，本文编号：2365691
本文链接：https://www.wllwen.com/jixiegongchenglunwen/2365691.html
