A Wavelet Transform-Based Approach for Joint Tracking in Gas Metal Arc Welding A new system was developed for joint tracking and control of the GMA welding process based on CCD sensors without an external light source

Effective tracking of the weld joint is crucial to the development of a high-performance control system for the gas metal arc welding (GMAW) process. This paper presents a new approach to ef- fectively track weld joints based on charge coupled device (CCD) sensors. Due to the presence of spatter, dust, and strong arc noises in welding environments, it has proved to be difficult to detect the weld joint from the CCD images captured in real time. In order to improve the robust- ness of weld joint tracking, this paper pre- sents a novel approach, based on the Bub- ble and M-band wavelet transform, for detecting the image edge of molten weld pools from the images captured during GMAW. The experimental results show that the effectiveness of the proposed method in detecting the edges of molten weld pools and identifying weld joints even when the welding images are pre- sented with much noise. Based on the weld joint identification, a PID control ap- proach is employed to manipulate the welding gun in order to produce a desired weld joint. The control experiments demonstrate that, based on the proposed joint tracking, the control of an S-shaped weld joint has been effectively delivered with good precision. BY J. X. XUE, L. L. ZHANG, Y. H. PENG, AND L. JIA

[1]  Oktay Alkin,et al.  Design of efficient M-band coders with linear-phase and perfect-reconstruction properties , 1995, IEEE Trans. Signal Process..

[2]  Zhang Jiankang,et al.  Orthogonal M-band compactly supported interpolating wavelet theory , 1999 .

[3]  Murat Kunt,et al.  High compression image coding via directional filtering , 1985 .

[4]  S.-J. Na,et al.  A Study on an Arc Sensor for Gas Metal Arc Welding of Horizontal Fillets An S-shaped fillet weld was successfully tracked using the quadratic curve-fitting method for sensing , 1991 .

[5]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Motoji Yamamoto,et al.  Application of fuzzy logic controller in the seam tracking of arc-welding robot , 1997, Proceedings of the IECON'97 23rd International Conference on Industrial Electronics, Control, and Instrumentation (Cat. No.97CH36066).

[7]  Suck-Joo Na,et al.  A study on vision sensors for seam tracking of height-varying weldment. Part 1: Mathematical model , 1997 .

[8]  Lale Akarun,et al.  Adaptive decimated median filters for image processing , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[9]  G. Goktug,et al.  Slit laser sensor guided real-time seam tracking arc welding robot system for non-uniform joint gaps , 2002, 2002 IEEE International Conference on Industrial Technology, 2002. IEEE ICIT '02..

[10]  Jianming Tao,et al.  Assessment of feedback variables for through the arc seam tracking in robotic gas metal arc welding , 1999, Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C).

[11]  L. Montefusco,et al.  Multiwavelet analysis and signal processing , 1998 .

[12]  John S. Baras,et al.  Properties of the multiscale maxima and zero-crossings representations , 1993, IEEE Trans. Signal Process..

[13]  Stéphane Mallat,et al.  Singularity detection and processing with wavelets , 1992, IEEE Trans. Inf. Theory.

[14]  Sanjit K. Mitra,et al.  Edge detection based on orientation distribution of gradient images , 1991, [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing.

[15]  Xiang-Gen Xia,et al.  On sampling theorem, wavelets, and wavelet transforms , 1993, IEEE Trans. Signal Process..

[16]  Y. Meyer.  Wavelets and Operators , 1993 .

[17]  Suck-Joo Na,et al.  A self-organizing fuzzy control approach to arc sensor for weld joint tracking in gas metal arc welding of butt joints , 1993 .

[18]  Radovan Kovacevic,et al.  Monitoring of weld joint penetration based on weld pool geometrical appearance , 1996 .

[19]  Ahmed H. Tewfik,et al.  Discrete orthogonal M-band wavelet decompositions , 1992, [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[20]  Mark J. Shensa,et al.  The discrete wavelet transform: wedding the a trous and Mallat algorithms , 1992, IEEE Trans. Signal Process..

[21]  Chen Nian.  A Visual Sensor Based Weld Seam Tracking Method for Precision Pulse TIG Welding , 2001 .

[22]  Peter N. Heller,et al.  Theory of regular M-band wavelet bases , 1993, IEEE Trans. Signal Process..

[23]  Hyungsuck Cho,et al.  A robust method for vision-based seam tracking in robotic arc welding , 1995, Proceedings of Tenth International Symposium on Intelligent Control.

[24]  Jae-Hoon Kim,et al.  Neural network-based recognition of navigation environment for intelligent shipyard welding robots , 2001, Proceedings 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems. Expanding the Societal Role of Robotics in the the Next Millennium (Cat. No.01CH37180).

[25]  Bülent Sankur,et al.  Multidirectional and multiscale edge detection via M-band wavelet transform , 1996, IEEE Trans. Image Process..

[26]  Hsing-Chia Kuo,et al.  An image tracking system for welded seams using fuzzy logic , 2002 .

[27]  Hirokazu Nomura,et al.  Automatic Real-Time Bead Height Control with Arc Sensor (Report 2) : Development of Multi-Pass MAG Arc Welding System , 1986 .