AI-based automatic weld path identification method

Abstract

The invention discloses an AI-based automatic weld path identification method. Three-dimensional model data of a workpiece needing to be welded is imported into a welding robot controller, and the imported three-dimensional model data of the workpiece comprises the overall structure of the workpiece needing to be welded and position information of a weld in the workpiece. Compared with a traditional automatic welding technology, the method has the advantages that a large amount of programming demonstration time can be shortened, and the production efficiency is improved; welding paths can be rapidly generated for different workpieces, and the link of designing complex tools is reduced; the problem of machining errors or position deviation existing in discharging of the workpiece can be well solved by using the method; and different welding modes such as fillet welding, vertical welding and lap welding can be automatically adjusted by applying the method for different welding types.

Description

technical field [0001] The invention relates to the field of robot welding, in particular to an AI-based automatic welding seam path identification method. Background technique [0002] At present, when using a welding robot for welding, it is necessary to manually teach the entire robot movement and weld trajectory. Therefore, when the structure of the welding workpiece or the shape and position of the weld change greatly, manual teaching will consume a lot of Due to the existence of processing errors, there are certain differences in the workpieces and welds of the same structure, and the use of the original manual teaching trajectory may affect the welding quality. At the same time, due to the lack of welding seam parameter detection equipment, traditional welding robots cannot adjust welding parameters according to different weld seams, and cannot effectively guarantee welding quality. In recent years, deep learning has made rapid progress in the fields of image recogni...

AI Technical Summary

Problems solved by technology

[0002] At present, when using a welding robot for welding, it is necessary to manually teach the entire robot movement and weld trajectory. Therefore, when the structure of the welding workpiece or the shape and position of the weld change greatly, manual teaching will consume a lot of labor cost and time cost; due to the existence of processing errors, there are certain differences between the same structural workpiece and weld seam, and the use of the original manual teaching trajectory may affect the welding quality

At the same time, due to the lack of welding seam parameter detection equipment, traditional welding robots cannot adjust welding parameters according to different weld seams, and cannot effectively guarantee welding quality

Images