机器人摆动弧焊纠偏的模糊神经网络控制系统

Fuzzy Neural Network Control System for Rectifying Deviation in Robotic Swaying Weld

讨论了机器人摆动弧焊的视觉跟踪控制系统。由CCD摄像机通过特殊设计的复合滤光系统摄取近弧区图象。图象处理后计算焊枪和焊缝之间的偏差。模糊神经网络用于控制纠偏机构的运动。实验结果证明了此方法的可行性。

A vision tracking control system in Robotic swaying weld is discussed in this paper. The near arc area image is catched by a CCD camera through a specially designed visible light compound filter system. The deviation between torch and seam is calculated after image processing. Fuzzy neural network is adopted to control the motion of rectifying mechanical device. The feasibility of this method is proved by the results of experiments.