3D Weld Seam Detection by Intersection Analysis and Hidden Seam Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The automation level of welding in industries like shipbuilding is low due to the complexity and time-consuming nature of robot programming, primarily because identifying welding seams is challenging.
Innovation Solution
A method and apparatus that identify welding seams in a three-dimensional model by tagging geometrical bodies, identifying intersection lines, eliminating hidden seams based on overlapping relationships, and determining joint edges, which facilitates automatic programming of welding robots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If manual programming methods are used for welding robots, then flexibility and adaptability are maintained, but automation level and productivity are reduced
Solution Approach 1:
The patent uses three-dimensional scanning to create a digital copy (point cloud data) of the actual workpiece geometry. This digital model replaces manual programming requirements, allowing the welding robot to automatically identify seams and generate welding paths from the scanned geometry, thereby increasing automation while managing complexity through digital replication
Solution Approach 2:
The patent replaces manual programming operations with automated image processing and geometric analysis algorithms. The system automatically processes point cloud data, identifies welding seams through image processing, and generates welding paths without mechanical manual intervention, substituting the programming process with automated computational methods
2Measurement precision
If traditional seam identification methods are used, then simplicity is maintained, but measurement precision and detection accuracy are reduced
Solution Approach 1:
The patent segments the complex task of seam identification into distinct processing stages: point cloud acquisition, point cloud processing, image generation, seam detection, and path planning. This segmentation allows each stage to be optimized independently, achieving high precision through specialized processing while managing overall system complexity through modular architecture
Solution Approach 2:
The patent transforms three-dimensional point cloud data into two-dimensional images for seam detection, then processes these images to identify welding seams. This dimensional transformation enables the use of efficient image processing algorithms to achieve precise seam identification while reducing the computational complexity of directly analyzing three-dimensional data
3Productivity
If comprehensive seam detection is performed, then productivity is improved, but loss of time in processing increases
Solution Approach 1:
The patent performs preliminary three-dimensional scanning and digital model creation before welding operations begin. By pre-processing the workpiece geometry and identifying all seams in advance, the system eliminates time-consuming on-site programming and setup, thereby increasing productivity while the one-time scanning investment offsets subsequent processing time
Solution Approach 2:
The patent creates a digital copy of the workpiece geometry through scanning, which can be processed and analyzed without affecting the actual welding timeline. This digital replica allows comprehensive seam detection and path planning to be performed offline, preventing time losses during actual welding operations
Data Source
AI summary
A method and an apparatus of identifying welding seams of a welding object. The method includes identifying intersection lines between geometrical bodies of the welding object in a three-dimensional model for the welding object, based on geometry of the geometrical bodies to form a collection of welding seams for the welding object. The method also includes eliminating hidden seams from the collection of welding seams for the welding object based on overlapping relationship among the welding seams to form a candidate seam list. Welding seams can be identified automatically and efficiently, and the method or the apparatus makes the automatic programming of the welding robot possible and thus facilitates use of a robot in welding huge and complex structures manufactured in a small batch.