NC AI develops welding AI for Hanwha Ocean

NC AI, a subsidiary of NCSoft, has secured a project from Hanwha Ocean to develop AI-powered autonomous welding technology for shipbuilding, addressing challenges like arc light and contamination. The solution combines a vision-based welding model and a robotic system, leveraging NC AI’s Vaetki Vision model for real-world application in commercial and special-purpose vessel production.
NC AI, the artificial intelligence arm of gaming company NCSoft, announced on Thursday it has won a project from Hanwha Ocean to develop AI-driven autonomous welding technology for shipbuilding. The system integrates a vision-based welding model with a collaborative robot designed to minimize human intervention while identifying and executing welding tasks under demanding conditions. The technology will be deployed at Hanwha Ocean’s commercial and special-purpose vessel production sites, where welding is one of the most complex processes. Challenges like arc light, sparks, fumes, outdoor environments, and lens contamination complicate stable image recognition for AI. NC AI is collaborating with Hanwha Ocean to incorporate real worksite data and engineer feedback into the development. The final system is intended for use in constructing Hanwha Ocean’s next-generation commercial ships and special-purpose vessels. NC AI’s proprietary Vaetki Vision, an industry-specific vision-language model set for release this year, will power the project. CEO Lee Yeon-soo stated that partnering with Hanwha Ocean—a leader in Korea’s shipbuilding sector—will demonstrate Vaetki’s scalability as a sovereign AI foundation model. The company aims to create robust vision recognition and autonomous control systems capable of overcoming dust and contamination in actual worksite conditions. This project marks a step toward advancing AI-driven automation in high-precision, high-stakes industries like shipbuilding.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.