Abstract: Welding robot is widely used in many kinds and working conditions of welding production in the current machinery manufacturing industry. It plays an essential role in the machinery manufacturing industry. At the moment, in most industries, welding robots still work by teaching and payback. When the welding object or conditions change, the robot cannot make corresponding adjustments in time, which makes the welding gun deviate from the weld center, resulting in the decline of welding quality. The realization of automatic and intelligent welding is the inevitable development trend in the future. The application of machine vision in the welding field will promote the transformation of welding technology from rigid welding automation to flexible welding intelligence. Welding automation and intelligence are intended to improve the working conditions and environment, reduce labor costs, and improve product quality. Robotic welding technology is known for its great efficiency and consistent quality. A four-step welding seam tracking system is suggested based on segmented scanning, filtering, feature points extraction, and path planning. Through the laser sensor installed at the end of the welding robot, the welding seam data is continuously collected in multiple segments in a segmented scanning manner. To improve the tracking accuracy, a combined filtering method is used to correct the data to reduce the effects of burrs, data distortion, and noise on the surface of the weldment. Then the feature points are collected, and the coordinate system is calibrated in order to identify the welding points. Finally, the spatial welding path is obtained by path planning. Two-dimensional type S and three-dimensional complex welding experimental investigations are carried out. The results show that the proposed method can form a complete welding path. The average errors of the two weldments are about 0.296 mm and 0.292 mm, respectively, which are close enough to fulfill the required accuracy of 0.5 mm. It shows that the proposed tracking method is effective and can provide a reference for the research of high-precision tracking and automatic welding.