Fillet weld positioning method based on point cloud geometric features

Abstract

The invention relates to the field of welding automation, in particular to a fillet weld positioning method based on point cloud geometric features, which comprises the following specific steps: S1, presetting the initial position of camera shooting; S2, adopting a multi-angle shooting result reconstruction model; S3, calculating a conversion matrix; S4, transferring to a robot coordinate system; S5, calculating the observation position and posture of each welding seam; S6, controlling the robot to move to the corresponding pose to shoot the point cloud; S7, obtaining the point cloud of the workpiece through preprocessing; S8, converting the point cloud into a robot coordinate system; S9, carrying out point cloud splicing; S10, generating a welding workpiece point cloud; S11, segmenting the point cloud cluster into a plurality of planes; S12, calculating a point cloud intersection line between faces; S13, calculating concavity and convexity of intersecting lines, and eliminating concave lines; S14, calculating a starting point, a key point and a normal vector of the line; and S15, generating and executing final welding seam information by the robot, and fusing point clouds of all the positions into a relatively complete workpiece point cloud model.

Description

technical field [0001] The invention relates to the field of welding automation, in particular to a fillet weld positioning method based on point cloud geometric features. Background technique [0002] Welding is a relatively basic technical link in the manufacturing industry and has a wide range of application scenarios. In the welding scenario of industrial robots, common automated welding methods include seam tracking and seam positioning. For example, in a welding seam tracking system and method based on laser sensors with Chinese patent number 201811282079.6 and an automatic welding seam tracking system based on polarized light illumination with Chinese patent application number 201811168970.7, seam tracking usually takes a long time to identify And positioning welds, and also requires the user to manually set the position of the weld, which reduces the efficiency of use and increases labor costs and difficulty in use. For example, a welding seam positioning device wi...

AI Technical Summary

Problems solved by technology

For example, a welding seam positioning device with a Chinese patent application number of 201711479984.6 and a welding method using a weld seam positioning device, and a welding positioning device with a Chinese patent application number of 201811377160.2 that is convenient for finding and locating welds depends on some specific positioning devices And equipment, the existence of these devices and equipment leads to higher use conditions and use costs of the welding system

Images