A novel BIM and vision-based robotic welding trajectory planning method for complex intersection curves

• Published in: Measurement : journal of the International Measurement Confederation Vol. 253; p. 117587
• Main Authors: Li, Tiejun, Meng, Shikang, Lu, Chaoyang, Wu, Yi, Liu, Jinyue
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.09.2025
• Subjects: BIM; Machine vision; Point cloud registration; Robotic welding systems; Welding seam extraction; Welding trajectory planning; Point cloud registration; Robotic welding systems; Welding seam extraction; Machine vision; Welding trajectory planning; BIM
• ISSN: 0263-2241
• DOI: 10.1016/j.measurement.2025.117587

•Fast and efficient weld feature extraction based on BIM.•A registration method based on the combination of axis direction vector and ICP.•Realize automatic welding trajectory planning in complex intersection curves welds.•Laser vision sensors track and correct the welding path. With the accelerated advancement of intelligent manufacturing, welding robots are increasingly utilized in industrial fields such as aerospace, shipbuilding, and heavy machinery. Traditional teaching-playback and offline programming methods are insufficient to meet the high-precision requirements for welding complex intersection curves. Additionally, machining and assembly errors in complex intersection curve components can cause deviations in the taught welding paths, leading to discrepancies from the intended trajectories and compromising welding quality. To address these challenges, we propose a novel building information modeling (BIM) and vision-based robotic welding trajectory planning method. First, a BIM-based algorithm extracts weld features from the workpiece. Then, a registration algorithm using axis direction vector features aligns the workpiece model with the measured point cloud. The weld information is transformed into the robot base coordinate system to determine the initial weld position. A laser vision sensor dynamically corrects the welding path in real time. Experimental results demonstrate that the proposed method effectively achieves weld feature extraction, path planning, and real-time tracking, maintaining a welding accuracy within 1 mm. This significantly enhances the reliability of robotic welding in complex manufacturing environments.