控制工程 ›› 2019, Vol. 26 ›› Issue (6): 1055-1059.

• 运动体控制系统 • 上一篇    下一篇

无标定视觉伺服机器人跟踪控制策略研究

  

  • 出版日期:2019-06-20 发布日期:2023-10-27

Research on Tracking Control Strategy of Uncalibrated Robot Visual Servo System

  • Online:2019-06-20 Published:2023-10-27

摘要: 针对无标定视觉伺服系统,eye-in-hand型结构,提出一种以图像特征为反馈的模糊控制策略,对平面内运动物体进行跟踪。应用Matlab和机器人工具箱Robotic Toolbox对控制策略进行仿真,并以三轴直角坐标系机器人、摄像机、计算机为基础,构建一种无标定视觉伺服机器人的运动跟踪系统。系统的工作原理如下,由摄像机采集图像,经图像处理目标定位,识别出待跟踪物体与当前摄像机末端的相对位置,通过模糊控制器对机器人的动作状态进行控制,实时反馈图像信息,纠正系统误差,最终达到稳定跟踪。仿真与实验验证了系统的可行性与实用性,为提高工业自动化生产线快速抓取提供了新的思路,对后续无标定伺服系统的研究做了铺垫。

关键词: 机器人, 视觉伺服, 图像特征, 运动跟踪, 无标定

Abstract: Aiming at the un-calibrated robot Visual Servo system of the eye in hand structure, a fuzzy control strategy with image feature as feedback is proposed to track the moving objects in the plane. The control strategy is simulated by Matlab and Robotic toolbox. The motion tracking system of an un-calibrated robot Visual Servo system is constructed based on three axis Cartesian coordinate system robot, camera and computer. The working principle of the system is as follows: the current image is collected by the camera, then the relative position of the object and the end of robot is measured by Matlab image processing, the action state of the robot is controlled by the fuzzy controller. According to the image real-time feedback from camera, constantly correct the deviation, and ultimately achieve stable tracking. The experiment and simulation experiment of physical crawling verify the feasibility and practicability of the system. It provides a new way to improve the rapid crawling of industrial automation production line, and paves the way for the follow up of un-calibrated servo system.

Key words: Robot, visual servo, image feature, motion tracking, un-calibration