基于双目视觉的目标识别与定位及机械臂的抓取研究

Research on Target Recognition and Location Based on Binocular Vision and Grasping of Manipulator

如今机械臂自动化生产线给制造业带来极大变革,为了提高机械臂在智能制造发展道路上面对复杂环境及复杂任务的自主性和功能性,加强机械臂对外界环境的感知能力,机器视觉技术应运而生。其中对于视觉技术的识别与定位可通过双目立体视觉技术来解决。该研究将可获取环境信息的视觉传感器与机械臂相互连接,对机械臂进行建模。随后搭建试验系统并进行相关标定,赋予机械臂目标识别和定位的功能,及对基于双目识别机器视觉技术的机械臂抓取进行研究。随后进行了双眼视觉机器人与三眼视觉机器人的算法效能比较,结果显示基于双眼视觉的机器人相比基于三眼视觉的机器人拥有更为灵活的移动精度、控制精度及工作效率。该研究改善了在复杂工作环境及复杂工作任务下,机械臂的自主调整能力和交互能力,为未来机械臂自主作业的研究提供方向,对工业、医疗等领域自动化水平的提升有显著意义。

Nowadays,the automatic production line of mechanical arm has brought great changes to the manufacturing industry.In order to improve the autonomy and functionality of mechanical arm in the face of complex environment and complex tasks on the development road of intelligent manufacturing,and to enhance the perception ability of mechanical arm to the external environment,machine vision technology emerged at the right moment.The recognition and localization of vision technology can be solved by binocular stereo vision technology.In this study,the visual sensor that can obtain environmental information is connected to the manipulator,and the manipulator is modeled.Then,the experimental system was set up and relevant calibration was carried out to give the robot arm the function of target recognition and positioning,and the robot arm grasping based on binocular recognition machine vision technology was studied.Then the algorithm efficiency of binocular vision robot and three-eye vision robot is compared.The results show that the binocular vision based robot has more flexible movement precision,control precision and work efficiency than the three-eye vision based robot.This study improves the autonomous adjustment ability and interaction ability of the manipulator in complex working environment and tasks,which provides a direction for future research on the autonomous operation of the manipulator,and has significant significance for the improvement of automation level in industry,medical and other fields.