首页> 外文会议>International Conference on Machine Learning and Cybernetics >Object Grabbing With A Five-Finger Robot Hand Using Two-Camera Vision
【24h】

Object Grabbing With A Five-Finger Robot Hand Using Two-Camera Vision

机译:使用双相机视觉用五指机器人手抓住对象

获取原文

摘要

This paper discusses the problem of visual-based object grabbing using a five-finger robotic hand connected to a 4-DOF manipulator on an intelligent mobile robot. The CamShift algorithm is used to analyze the dynamic vision acquired by a pair of stereo cameras in three-dimension space. Afterwards, inverse and forward kinematics of manipulator and fingers are performed to drive the robot hand to appropriate position and grab a cup. In addition, a coordinate estimation correction method was proposed to improve the accuracy of the calculated position. The proposed method is validated in several experiments.
机译:本文讨论了使用连接到智能移动机器人的4-DOF机器人的五指机器人抓取基于视觉对象抓取的问题。 CAMShift算法用于分析三维空间中一对立体声相机获取的动态视觉。之后,执行操纵器和手指的反向和向前运动学,以将机器人手驱动到适当的位置并抓住杯子。另外,提出了一种坐标估计校正方法来提高计算位置的准确性。提出的方法在几个实验中验证。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号