NC AI Secures Project for "Development of Autonomous Welding Model"

NC's artificial intelligence (AI) specialist division, NC AI, announced on June 4 that it has officially launched a collaboration system after winning the final contract for Hanwha Ocean's project to develop a "vision recognition-based dedicated welding model and a collaborative robot-based autonomous welding model."


NC AI logo. Provided by NC AI

NC AI logo. Provided by NC AI

View original image

This project aims to integrate advanced AI vision recognition and precise robot control technologies into the welding process, which is a core step in shipbuilding that has traditionally relied on the expertise of skilled workers. NC AI and Hanwha Ocean are working toward developing an autonomous welding physical AI solution, where robots independently identify and assess welding areas and perform optimal welds in real time.


NC AI stated that it is conducting on-site, hands-on research that incorporates actual worksite data and engineers’ feedback from Hanwha Ocean in real time. Due to the nature of shipyard welding processes—featuring intense arc light and sparks, real-time welding fumes, and outdoor or harsh work environments that can contaminate camera lenses—the conditions are challenging for vision recognition AI. To address this, NC AI explained that it aims to perfect "shipbuilding-specialized vision recognition technology" capable of extracting geometric weld lines and detecting welding defects in real time, even amid strong noise and contamination.


For this autonomous welding technology, NC AI plans to apply its next-generation industrial vision-language model (VLM), "VAETKI Vision." VAETKI Vision is an industry-specialized VLM that integrates visual environmental data and text instructions, simultaneously processing vision, language, and action to control the robot’s physical movements—thus expanding into a Vision-Language-Action (VLA) model. Without complex coding, workers can give spoken or written commands, and the collaborative robot will use VAETKI Vision to visually analyze the shape of the welding target and the condition of the weld line in real time, autonomously deriving appropriate actions.



Yeon-Soo Lee, CEO of NC AI, stated, "We are committed to developing world-class physical AI models that can be deployed in real shipbuilding processes by successfully overcoming dust and contamination in the field through advanced vision recognition technology and autonomous control models."


This content was produced with the assistance of AI translation services.

© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.

Today’s Briefing