基于视觉感知的机械臂运动控制研究

发布时间：2019-07-05 15:57

【摘要】：随着科学技术与现代工业的飞速发展,机器人在各个领域扮演着越来越重要的角色,传统的机器人适应与感知环境能力较弱,已不能满足人类的需求。近年来,视觉技术的发展日新月异,机器视觉让机器人拥有了更加智能的环境适应能力,视觉系统以及视觉感知技术成为智能机器人的重要组成部分。具有视觉感知能力的机器人成为当今社会科研工作者的热门研究方向。本文对具有双目视觉感知的机器人物品抓取系统进行了深入理论研究和实验验证。本文基于Linux开发环境,设计实现了基于双目视觉感知的物品抓取机器人系统。嵌入式开发平台通过对双目相机采集的目标物体图像进行处理,获得目标物体的空间三维坐标信息,并传递至机械臂的控制端,机械臂控制机构根据设定的机械臂关节调整策略控制各个关节转动完成对目标物体的抓取作业。本文研究内容主要分为以下三个方面:(1)研究机器人的硬件构成,完成机器人的结构设计。主要包括研究机器人各个模块的控制原理与通过相机标定获取相机的内外参数与畸变系数。(2)研究相机成像模型,完成双目测距与手眼关系标定,获得目标物体相对于机器人抓取机构的三维坐标。首先通过极线约束原理对相机进行立体校正,使用SGBM算法对相机采集的图像对进立体匹配,经深度恢复与手眼关系标定,得到目标物体空间三维坐标。(3)研究机器人抓取机构的正逆运动学问题,设计机械臂关节自动调整策略,验证机械臂的结构合理性。
文内图片：

图片说明：机器人实物图
[Abstract]:With the rapid development of science and technology and modern industry, the robot plays a more and more important role in various fields. In recent years, the development of vision technology is changing with each passing day, and the machine vision has made the robot more intelligent, and the vision system and the visual perception technology become an important part of the intelligent robot. The robot with the ability of visual perception becomes the hot research direction of the researchers in the present society. In this paper, an in-depth theoretical research and experimental verification of a robot article grasping system with binocular vision perception is carried out. In this paper, based on the development environment of Linux, the robot system based on binocular vision perception is designed. the embedded development platform processes the target object image acquired by the binocular camera, obtains the spatial three-dimensional coordinate information of the target object, and transmits the space three-dimensional coordinate information to the control end of the mechanical arm, And the mechanical arm control mechanism controls each joint rotation to finish the grabbing operation on the target object according to the set mechanical arm joint adjustment strategy. The content of this paper is mainly divided into the following three aspects: (1) The hardware structure of the research robot is studied, and the structure design of the robot is completed. The control principle of each module of the robot is mainly studied, and the internal and external parameters and the distortion coefficient of the camera are obtained by the camera calibration. And (2) the camera imaging model is researched, the binocular distance measurement and the hand-eye relation calibration are finished, and the three-dimensional coordinates of the target object with respect to the robot gripping mechanism are obtained. Firstly, the camera is subjected to three-dimensional correction through a polar line constraint principle, the image acquired by the camera is matched with the stereo matching by using the SGBM algorithm, and the three-dimensional coordinate of the target object space is obtained through the depth recovery and the hand-eye relation calibration. (3) To study the inverse kinematics problem of the robot grasping mechanism, and to design the automatic adjustment strategy of the mechanical arm joint to verify the structural rationality of the mechanical arm.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41;TP241

【参考文献】