Abstract: In order to realize the curve weld seams automatic tracking with welding robot, a simple real-time posture adjustment strategy and coordinate visual tracking and visual servoing control of the welding torch movement method is proposed in this paper. The mathematical model of expected position in the process of robot motion is established. Also, a fuzzy visual servo controller based on the?up-down structure is designed. Through the establishment of the relational model which depends on coordinates deviation of weld feature points pixel and end shaft rotation angle, to determine the size of fuzzy domain of discourse dynamically. While only by adjusting the amount of end shaft rotation, it ensures that the image feature point always exists in the camera view on the basis of the robot with ideal position. Simulation results of curve weld automatic tracking verify the effectiveness of the proposed method.

Key words: welding robot, posture adjustment, tracking curve weld seams, visual servoing control

摘要： 为实现焊接机器人对曲线焊缝的自动跟踪，提出一种简便的位姿实时调整策略和协调视觉跟踪与机器人运动的视觉伺服控制方法。建立了曲线焊缝视觉跟踪过程中焊接机器人期望位姿的数学模型；设计了一种上下层结构的模糊视觉伺服控制器，通过建立焊缝特征点像素坐标偏差与末端轴旋转角度之间的关系模型，动态确定模糊论域的大小，在机器人期望位姿的基础上仅仅通过调整末端轴的旋转量来保证图像特征点始终存在于相机视场内。通过模拟焊接机器人自动跟踪曲线焊缝的实验，验证了所提策略与方法的有效性。

关键词: 焊接机器人, 位姿调整, 曲线焊缝跟踪, 视觉伺服控制