AI robots become the 'eyes' of other robots, penetrating manufacturing sites with 'MAX'
Translated from Korean, summarized and contextualized by DistantNews.
At a glance
- HD Hyundai Heavy Industries is using AI-powered robots to automate the production of 'lugs,' essential components for lifting ship blocks.
- The 'LUG Autonomous Manufacturing System,' supported by the Ministry of Trade, Industry and Energy's 'MAX' initiative, has increased lug production by 87.5%.
- This initiative aims to integrate AI into manufacturing across various industries, moving towards fully autonomous 'dark factories'.
At HD Hyundai Heavy Industries' medium ship business division in Ulsan, South Korea, a trio of yellow robotic arms are diligently working. These robots have divided the labor for producing 'lugs' โ crucial steel components used to lift ship blocks. The first robot precisely places a steel part, the second transfers it to a welding station, and the third initiates the welding process, sending sparks and smoke into the air.
Vision AI is equipped with robots that photograph parts to identify their exact location, and transmit that information to robots without eyes to continue parts production such as welding.
This sophisticated operation is powered by a vision AI system. Senior manager Hwang Sang-min explained that the AI-equipped robots accurately identify component positions and transmit this data to 'blind' robots, enabling them to continue tasks like welding. These lugs, though relatively small, are indispensable for connecting lifting equipment to ship blocks, forming essential parts of the massive structures built in the shipyard. The 'LUG Autonomous Manufacturing System' at HD Hyundai Heavy Industries is a prime example of the Ministry of Trade, Industry and Energy's 'Manufacturing AI Transformation (MAX)' initiative being applied in real production environments.
Since implementing this system, lug production has surged by an impressive 87.5%. The MAX initiative is actively promoting the integration of AI into manufacturing across different sectors. At POSCO's raw material plant, robots are being developed to inspect belt conveyors, analyzing signs of malfunction through sound and heat, with the goal of replacing faulty rollers without halting operations. Similarly, at POSCO's Pohang Steelworks, quadrupedal robots are patrolling high-temperature equipment near blast furnaces, monitoring temperature and gas leaks. POSCO is also developing mobile robotic arms to take over tasks like temperature measurement and sample collection, which were previously performed by humans in hazardous, high-heat environments.
The lug autonomous manufacturing system is an example of the Ministry of Trade, Industry and Energy's Manufacturing AI Transformation (MAX) support being applied to the production site.
At EcoPro's battery material production site in Pohang, AI is being used to predict quality, while robots conduct equipment inspections. Autonomous Mobile Robots (AMRs) named 'Tifoy' navigate around calcination furnaces, using acoustic, thermal imaging, and video sensors to detect pipe leaks and equipment anomalies. This automation of routine inspections, previously done by humans in environments with high temperatures and dust, allows robots to handle approximately 70% of the roughly 200 inspection items.
The robots inspect belt conveyors for raw material plants at POSCO, analyzing signs of malfunction through sound and heat, and aim to replace faulty rollers without stopping equipment operation.
The benefits of these national projects extend beyond financial support. They accelerate AI adoption and broaden its application scope. For instance, POSCO's project involves 10 entities, including research institutes, robot manufacturers, and universities. POSCO provides real steelmaking processes and data, while participating institutions collaboratively develop robots, sensors, and AI technologies for early fault detection, with field performance validation. EcoPro BM's project similarly involves six organizations, including research institutes and IT firms, jointly developing data platforms, AI training data, and quality prediction models. The verified technologies are planned for expansion to other processes and industries. POSCO envisions integrating robots, sensors, and digital twins into a unified operating system for modular products applicable to industries with similar equipment, such as cement plants and thermal power plants. HD Hyundai Heavy Industries plans to standardize its jointly developed system for use by partner companies and smaller shipbuilders, while offering consulting to collaborators interested in applying lug autonomous manufacturing technology to other processes. EcoPro BM aims to expand its quality prediction technology into a fully autonomous manufacturing line encompassing production, quality, equipment, and safety.
The quadrupedal robot patrols around high-temperature equipment near the blast furnace, checking for temperature and gas leaks.
Originally published by Hankyoreh in Korean. Translated, summarized, and contextualized by our editorial team with added local perspective. Read our editorial standards.