MakinaRocks与HD韩国造船完成船厂焊接机器人AI异常检测模型开发

Physical artificial intelligence (AI) company MakinaRocks (477850) said on the 4th that it has completed development of an AI-based anomaly detection model for shipyard welding robots, in collaboration with HD Korea Shipbuilding & Offshore Engineering (009540).
The project is a follow-up to the "Memorandum of Understanding (MOU) on Using AI for Robot Process Efficiency" signed by the two companies last year. It was pursued with the goal of rapidly applying AI to shipyard sites.
In shipbuilding, welding is a core process that determines the structural stability of a ship. A minor malfunction of a robot can lead to large-scale rework and losses in upstream and downstream processes. This is the backdrop against which anomaly detection technology, which detects a robot's condition in advance, has emerged as a variable in shipyard productivity and profitability.
Considering the characteristics of industrial sites where failure history data is limited, this task was pursued based on normal operation data. Through analysis of experimental facility data, MakinaRocks identified the main causes and patterns of anomaly signs in welding robots. It then refined the model using actual data from 12 welding robots used at the shipyard.
The model was verified along with a monitoring function that visualizes anomaly scores and key signals. The company plans to review the possibility of linking it with on-site operating systems in the future.
"Physical AI, which overcomes the data limitations of industrial sites and puts AI into operation, is a technology trend that global manufacturing companies cannot avoid," MakinaRocks CEO Yoon Sung-ho said. "Based on this collaboration, we will present a technological milestone for physical AI that can be applied to shipbuilding industry sites."