RESEARCH ON MACHINE VISION METHODS FOR ROBOT MANIPULATOR LOCATION
To deal with the problem of robot manipulator’s accuracy of location, a method based on stereo vision is proposed. The method includes three sections: first, set up a reference frame between search area and the target-image based on area-based matching; second, feature extraction on the image; third, calculate errors between initial and desired joints’ position based on the robot kinematics. Motoman which has 6DOFs is used for test. The results show that this method can correctly and quickly find out the features of target, and then the controller drives these errors to zero. It can achieve high rates of tracking and locating.
Robot manipulator machine vision feature extraction tracking
Liao Wanhui Li Lin
Robotics Research Laboratory,School of mechanical and auto engineering,South China;University of Technology Guangzhou,Guangdong,China
国际会议
2008年拟人系统国际会议(2008 International Conference on Humanized Systems )(ICHS’08)
北京
英文
2008-10-18(万方平台首次上网日期,不代表论文的发表时间)