Abstract: In order to realize service robot flexibly self-location and accurately grasping objects, this paper proposes a self-location and grasping algorithm based on vision and Radio Frequency Identification(RFID). It gives the grid environment, uses RFID to determine the initial position, route and direction of the robot. This paper uses visual system to get the space coordinates of objects, converts them to arm coordinate system, models with the improved D-H model for arm, and gives manipulator inverse solution graping algorithm. Experimental results show that this algorithm can make the robot location success rate reach 76.7%, and graping success rate is 90%.

Key words: service robo, Radio Frequency Identification(RFID), self-location, grasping, manipulator, binocular vision

摘要： 为实现机器人灵活的自定位，并使其准确地抓取物体，提出一种基于视觉与无线射频识别(RFID)技术的机器人自定位抓取算法。构建网格化环境，利用RFID技术确定机器人的初始位置、行进路线和方向，使用视觉系统获取物体的空间坐标，将其转换到手臂坐标系，采用改进的D-H模型对手臂进行建模，并给出机械臂逆解抓取算法。实验结果表明，该算法使得机器人定位的成功率达到76.7%，抓取成功率高达90%。

关键词: 服务机器人, 无线射频识别, 自定位, 抓取, 机械臂, 双目视觉

CLC Number: 

• TP18