A Robust Weld Seam Tracking Method by Using Intensity-Modulated Structured Light
Wenjun Shao, Yu Huang
Abstract
The seam tracking method by using structured light is widely used in the automatic welding process. Most of the noise can be filtered out by a narrow-band optical filter whose center wavelength is the same as the structured light. However, it still suffers from light noise caused by sparks and fumes when the structured light is getting close to the welding pool. This limitation of the structured-light based method is inherent because light noise caused by the welding process has a wide spectrum and couldn't filter out by the optical filter. To circumvent this problem, we proposed a seam tracking method that can modulate the intensity of the structured light at a high frequency by an acoustic-optic modulator (AOM) and extract the modulated frequency from multiple seam images by Fourier transformation. As the frequency of noises such as sparks is random and different from the modulated frequency, the proposed method can remove these noises in the frequency domain directly. We demonstrated the robustness of this method by detecting seam in noisy welding conditions.