Application of Intelligent Steel Bar Welding Robot Based on Image Tracking Technology

DOI: https://doi.org/10.62517/jcte.202506414

Author(s)

Cai Yu*, Yang Wen, Guo Bing, Yin Ruiyang, Cheng Shufan

Affiliation(s)

Hubei province road &bridge Co. Ltd., Wuhan, China *Corresponding Author

Abstract

Most existing steel bar welding robots adopt a pre-programmed operation mode and lack dynamic adjustment capabilities. To promote the upgrading of welding robots from mechanical operation to intelligent operation, an intelligent steel reinforcement cage welding robot was developed based on image tracking technology through the optimization of tooling fixtures and integration of core systems. The robot utilizes a V-shaped integrated supporting and pressing tooling fixture and a three-axis linked gantry positioning system, which enables control of the longitudinal, transverse, and height deviations of upper-layer steel bars within 1.0 mm, with a repeat positioning deviation of less than 0.5 mm. An industrial-grade 3D structured light camera and a weld seam tracking system are integrated at the end of the robotic arm. A coordinate mapping between the camera and the robotic arm is established using the chessboard calibration method, forming a closed-loop technical process of photographing, recognition-3D modelling judgment, and dynamic correction, with a calibration error of < 0.1 mm. Tests show that the total time for a single welding spot of the robot, from positioning to welding, is ≤ 13.6 seconds, which is 45.6% more efficient than traditional pre-programmed robots. The welding qualification rate reaches over 98.7%. It can meet the high-efficiency and high-precision welding needs of mass production of steel reinforcement cages in intelligent beam yards.

Keywords

Welding Robot; Weld Seam; 3D Structured Light Camera; Tooling Fixture; Dynamic Correction; Intelligent Beam Yard

References

[1] Palega, M., Salwin, M., Chmielewski, M., et al. (2025) An Innovative Approach to Welding Robot Operator Reliability Analysis Using Extended TESEO and HEART Methods. Scientific Reports, 15(1), 39123. [2] Ahmed, F., Barman, M., Shaik, F., et al. (2025) Optimizing ABB MIG Welding Robot Through Polynomial Trajectory Planning and Artificial Intelligence Integration. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 239(18), 7504-7521. [3] Zhu, C., Wang, Z., Zhu, Z., et al. (2025) Design and Modeling of Pipeline Welding Robot Based on Pieper Criterion and Construction of Integrated Intelligent Welding System. Welding in the World, 69(9), 1-15. [4] Chi, P., Wang, Z., Liao, H., et al. (2025) An Integrated Calibration and 3D Reconstruction Method for Humanoid Welding Robots. Measurement, 253(PD), 117748. [5] Gong, Z., Yuhang, Z., Shuaihua, T., et al. (2021) A Novel Seam Tracking Technique with a Four-Step Method and Experimental Investigation of Robotic Welding Oriented to Complex Welding Seam. Sensors, 21(9), 3067. [6] Liu, W.Q., Li, D.B., Ding, S.C., et al. (2025) Research on Automated Welding Robot Technology for Reinforcement Lap Joints of Stiffened Steel Structures in Extra-Large Metro Depots. Welding Technology, 54(07), 102-108. (in Chinese) [7] Wu, Y.Y., Zhang, G., Lin, Q.X., et al. (2021) Research Review and Prospect of Characteristic Parameters Prediction Methods for Welding Robot. Machine Tool ＆ Hydraulics, 49(15), 168-173+199. (in Chinese) [8] Wang, P., Zhang, D., Lu, B. (2020) Robust Fuzzy Sliding Mode Control Based on Low Pass Filter for the Welding Robot with Dynamic Uncertainty. The Industrial Robot, 47(1), 111-120. [9] Yao, S., Xue, L., Huang, J., et al. (2025) Mobile Welding Robots Under Special Working Conditions: A Review. The International Journal of Advanced Manufacturing Technology, 138(9), 1-17.