计算机工程 ›› 2009, Vol. 35 ›› Issue (18): 200-202.doi: 10.3969/j.issn.1000-3428.2009.18.070

• 人工智能及识别技术 • 上一篇    下一篇

基于目标跟踪的并联机器人视觉位姿检测

陈建林1,丁永生1,2,郝矿荣1,张淑平1   

  1. (1. 东华大学信息科学与技术学院,上海 201620;2. 东华大学数字化纺织服装技术教育部工程研究中心,上海 201620)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-09-20 发布日期:2009-09-20

Vision Pose Measurement for Parallel Robot Based on Object Tracking

CHEN Jian-lin1, DING Yong-sheng1,2, HAO Kuang-rong1, ZHANG Shu-ping1   

  1. (1. College of Information Sciences and Technology, Donghua University, Shanghai 201620;2. Engineering Research Center of Digitized Textile & Fashion Technology, Ministry of Education, Donghua University, Shanghai 201620)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-09-20 Published:2009-09-20

摘要: 建立一种基于视觉的并联机器人位姿检测系统框架,包括图像采集、图像处理、位姿检测、参数反馈4个部分。使用单目摄像头采集图像,以二自由度冗余机器人为控制对象,利用Haar特征提取对目标进行粗跟踪。进一步获得目标上特定的几个特征点,基于平行不变性原理,得到机器人末端操作器的实际位姿参数。通过求解机器人的逆运动学方程,得到电机的控制参数。实验和仿真验证了该系统的可行性。

关键词: Haar特征, 二自由度并联机器人, 位姿检测

Abstract: This paper presents a vision based feedback system framework of position and orientation measurement for parallel robot. It is composed of image acquisition, camera calibration, position and orientation measurement, and parameter feedback. Single camera is used for image acquisition. A kind of 2-DOF parallel robot is used as control object, and Haar-like feature detection is presented as a method of rough object tracking. In order to acquire the real parameters based on invariants of trapezia, the special feature points on the object are determined. An inverse-kinematics of robot is used to get the control signals of every motor. And the feasibility of the system is proved by simulation experiments.

Key words: Haar feature, 2-DOF parallel robot, pose measurement

中图分类号: