Monitoring Welding Torch Position and Posture Using Reversed Electrode Images – Part I: Establishment of the REI-TPA Model

Abstract

Sensing, modeling, and control are the key technologies of intelligent welding manufacturing. The position and attitude of the welding torch relative to the weld seam affects the quality of the produced weld directly, which is the basis of control and offline programming of the welding robot. Based on the existing research of reversed electrode image (REI) on monitoring of weld pool surface and penetration state, this work focused on the application of REI on monitoring of welding torch position and attitude (TPA). In this paper, a REI-TPA model was established to quantitively relate REI to TPA by considering the surface of the weld pool as a spherical mirror. With the REI-TPA model, the offset distance and deflection angle of the welding torch relative to the correct position and attitude can be calculated. The calculation is based on the measurement of the relative distance between the tip of REI and the electrode in a passive visual image of the weld pool, arc length, and weld pool geometry. This model can be applied to the real-time control of TPA, which is a supplement to the application of welding passive visual image and an extension to multi-information acquisition and processing methods in the robotic welding process.