Abstract: The welding positioning method for the support box cushion block of the container hold cover mainly relies on manual visual judgment of the approximate position of the welding hole, and then combines experience to determine the welding point. Due to the lack of precise capture of the circular features of the welding holes, the positioning accuracy is not high. A laser induced automatic positioning method for the installation and welding of container deck cover support box cushion blocks is proposed. Laser lines are projected onto the welding surface of the support box cushion block through structured light projection. The deformed laser stripes are collected by an industrial camera and the point cloud coordinates are calculated based on the principle of triangulation. Outliers in the welding point cloud data are removed using statistical filtering algorithms. The principal component analysis method is used to perform covariance decomposition on the local neighborhood of the point cloud, and the normal vector is determined in the direction of the minimum eigenvalue. The RANSAC algorithm is used to fit the straight line model of the weld seam, achieving the extraction of the centerline of the weld seam of the support box cushion block of the container hold cover. Extract points from the point cloud that conform to the cylindrical edge features using a random Hough circle transformation operation. Project the center of the detected bolt hole onto the pre extracted centerline of the weld seam to achieve welding positioning. In the experiment, the positioning accuracy of the proposed method was tested. The final test results show that when using the proposed method for welding positioning, the average overlap between the pad welding holes and the reference benchmark reaches 0.87, which has a relatively ideal positioning effect.