Deposition height detection and feature point extraction in robotic GTA-based additive manufacturing using passive vision sensing

摘要

Automatic detection of deposition height is one of the key technologies for metal components fabricated in robotic gas tungsten arc (GTA) based additive manufacturing. In this research, the deposition height, defined as the tungsten tip to the top layer distance, is monitored by a passive vision sensor consisting of a camera and optical filters. Characteristic parameters of molten pool tail are extracted by designed image processing algorithms. An innovative tracking algorithm is developed to determine the location of the solid-liquid separation point at the molten pool tail by means of continuous images. Finally, the detection approach is verified through depositing multi-layer single-pass parts, proving the excellent effectiveness and anti-interference ability of the tracking algorithm in robotic GTA-based additive manufacturing.