Enhancing weld quality of novel robotic-arm arc welding: Vision-based monitoring, real-time control seam tracking

被引:2
|
作者
Sharma, Aman [1 ]
Chaturvedi, Rishabh [1 ]
Sharma, Kamal [1 ]
Binhowimal, Saad Abrahim [2 ]
Giri, Jayant [3 ,4 ,5 ]
Sathish, T. [6 ]
机构
[1] GLA Univ, Dept Mech Engn, Mathura 281406, India
[2] King Abdulaziz City Sci & Technol, Adv Mfg Technol Inst, Riyadh 11442, Saudi Arabia
[3] Yeshwantrao Chavan Coll Engn, Dept Mech Engn, Nagpur, India
[4] Lovely Profess Univ, Div Res & Dev, Phagwara, India
[5] Chitkara Univ, Chitkara Univ Inst Engn & Technol, Ctr Res Impact & Outcome, Rajpura 140401, Punjab, India
[6] SIMATS, Saveetha Sch Engn, Chennai 602105, Tamil Nadu, India
关键词
Seam tracking; Robotic-Arm Arc Welding (RAAW); Aristo Robot; Accuracy; Real-time monitoring; Fault detection and diagnostics (FDD); Vision system; Hand-eye calibration; SYSTEM;
D O I
10.1016/j.asej.2024.103109
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
A specially developed vision sensor seam tracking system is installed to improve the welding quality for RoboticArm Arc Welding (RAAW). New software is used for welding seam monitoring to examine different features of robotic arm arc welding. The software system comprises several components: welding voltage regulation, analytical parameters, image capture or analysis using enhanced algorithms, weld experts' database, robot coordination, and trajectory operations management. Experimentation is carried out to ensure the quality and reliability of various traditional welding methods and to evaluate the effectiveness of an established seamtracking process. The results indicate that seam tracking is accurate for most welding processes. However, it is noted that most image-learning algorithms have limited adaptability and capacity during welding. A vision sensor-based algorithm for adaptive extraction functionality is proposed to solve this issue. Conventional welding seams are divided into laser strip pictures into continuous and discontinuous weld seams. The vision system then processes the captured data to convert the pixel coordinates into real-world coordinates, often using hand-eye calibration techniques to align with the robot's coordinate system. The vision system captures data and processes it to convert pixel coordinates into real-world coordinates. The alignment process with the coordinate system of the robot is done using hand-eye calibration techniques. The Exponentially Weighted Moving Average (EWMA) control panel is used to diagnose and identify faults, keeping an eye on anomalous deviations from the extracted slurry profile while welding. To achieve a 96 % seam extraction ratio and +/- 0.50 mm seam tracking errors, the system uses an Aristo six-axis robot, a custom vision sensor, and seam-tracking software. The system demonstrates accuracy in industrial welding processes, with maximum and average seam tracking errors of 0.56 mm and 0.48 mm for continuous seams and 0.36 mm and 0.30 mm for non-continuous seams.
引用
收藏
页数:17
相关论文
共 42 条
  • [31] Vision-based real-time monitoring of extrusion additive manufacturing processes for automatic manufacturing error detection
    Charalampous, Paschalis
    Kostavelis, Ioannis
    Kopsacheilis, Charalampos
    Tzovaras, Dimitrios
    INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY, 2021, 115 (11-12): : 3859 - 3872
  • [32] Vision-based real-time structural vibration measurement through deep-learning-based detection and tracking methods
    Pan, Xiao
    Yang, T. Y.
    Xiao, Yifei
    Yao, Hongcan
    Adeli, Hojjat
    ENGINEERING STRUCTURES, 2023, 281
  • [33] Monocular vision-based real-time target recognition and tracking for autonomously landing an UAV in a cluttered shipboard environment
    Lin, Shanggang
    Garratt, Matthew A.
    Lambert, Andrew J.
    AUTONOMOUS ROBOTS, 2017, 41 (04) : 881 - 901
  • [34] A novel vision-based real-time method for evaluating postural risk factors associated with musculoskeletal disorders
    Li, Li
    Martin, Tara
    Xu, Xu
    APPLIED ERGONOMICS, 2020, 87
  • [35] Real-time plant phenomics under robotic farming setup: A vision-based platform for complex plant phenotyping tasks
    Arunachalam, Ajay
    Andreasson, Henrik
    COMPUTERS & ELECTRICAL ENGINEERING, 2021, 92
  • [36] Real-time monitoring of abnormal conditions based on Fuzzy Kohonen clustering network in gas metal arc welding
    Gao J.
    Wu C.
    Hu J.
    Frontiers of Materials Science in China, 2007, 1 (2): : 134 - 139
  • [37] Development of a real-time laser-based machine vision system to monitor and control welding processes
    Huang, Wei
    Kovacevic, Radovan
    INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY, 2012, 63 (1-4): : 235 - 248
  • [38] Development of a real-time laser-based machine vision system to monitor and control welding processes
    Wei Huang
    Radovan Kovacevic
    The International Journal of Advanced Manufacturing Technology, 2012, 63 : 235 - 248
  • [39] Real-time nondestructive monitoring of Common Carp Fish freshness using robust vision-based intelligent modeling approaches
    Taheri-Garavand, Amin
    Fatahi, Soodabeh
    Banan, Ashkan
    Makino, Yoshio
    COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2019, 159 : 16 - 27
  • [40] A real-time obstacle avoidance and path tracking strategy for a mobile robot using machine-learning and vision-based approach
    Singh, Rajmeet
    Bera, Tarun Kumar
    Chatti, Nizar
    SIMULATION-TRANSACTIONS OF THE SOCIETY FOR MODELING AND SIMULATION INTERNATIONAL, 2022, 98 (09): : 789 - 805