Design of a directional feature-driven weld target tracking algorithm.

In order to solve the problems such as unique and strange working space which is not easy to weld, inconvenient operation of teaching reproduction working mode, poor tracking effect caused by environmental interference of tracking accuracy of welding seam, a welding target tracking algorithm based on direction feature was designed. The algorithm has good accuracy, strong stability and is suitable for field welding operations. The orientation feature-driven welding seam target tracking algorithm proposed in this paper processes and calculates the images and other data captured by the camera to obtain the error of welding seam tracking and carry out follow-up tracking operations, which can complete welding seam tracking more accurately and reliably. The algorithm proposed in this paper adapts to the image features of V-groove under the irradiation of auxiliary light source, and enables the camera and welding torch to find the weld coordinates more accurately under the interference of arc light, splash and other noises. The results of welding experiment in the tank show that the system has good stability, high accuracy and good weld forming quality, which is within 0.12mm and meets the process requirements of pipeline welding.