In general,two dimensional data sensors are used in the field of intelligent welding,where tracking of the weld seam area can lead to large tracking errors due to welding gun vibration.In this paper,a laser weld seam automatic tracking method for laser weld seam based on 3D laser scanning sensor is proposed,applying the 3D laser scanning sensor to obtain the complete 3D position data of laser weld seam.The collected weld position information is processed by the global optimal solution to extract the relevant features of the laser weld.The distance features related to laser weld seam tracking is classified.And a fuzzy tracking algorithm is designed to establish the target edge feature model using the basic universe method,to adjust the membership parameters of laser weld seam,and to achieve target fusion feature classification.On the basis of laser weld feature classification,the variable universe theory is used to improve the accuracy of welding tracking control and the efficiency of weld shape detection,to complete the control tracking of the weld area,and to realize unmanned welding.The experimental results show that this method has a significant effect in reducing the tracking error.