DEEPEX, Doosan Robotics Develop “AI Welding” for Nuclear Power Plant Manufacturing Sites
TECHWORLD ·
✦ AI Summary
DEEPEX and Doosan Robotics said on the 6th that they will pursue the development of an AI-based intelligent welding robot solution for nuclear power plant manufacturing sites.
Doosan Robotics, DEEPEX, SAIGE, and the Industry-Academia Cooperation Foundation of Changwon National University were selected for the Ministry of Trade, Industry and Energy’s “2026 Robot Industry Technology Development Project” and will jointly develop an AI welding solution for nuclear power plant manufacturing sites.
Demonstrations will then be carried out in stages at Doosan Enerbility nuclear power plant equipment manufacturing sites during the project period, with the goal of completing quality verification for nuclear power plant equipment in 2029.
DEEPEX and Doosan Robotics said on the 6th that they will pursue the development of an AI-based intelligent welding robot solution for nuclear power plant manufacturing sites. The purpose of the project is to address the shortage of skilled welders.
DEEPEX and Doosan Robotics have been selected for the Ministry of Trade, Industry and Energy’s “2026 Robot Industry Technology Development Project.” DEEPEX, Doosan Robotics, SAIGE, and the Industry-Academia Cooperation Foundation of Changwon National University will jointly develop an AI welding solution to be applied at nuclear power plant manufacturing sites. The four organizations signed an MOU on July 21 to carry out the project and established a joint technology development cooperation framework.
The core of the project is to combine vision AI and collaborative robot technology to make the welding process intelligent. Development work includes AI recognition of welding targets and paths, AI path correction, analysis of vision and thermal imaging data collected during welding, and technology to determine welding conditions and abnormalities.
The organizations plan to implement the technologies they develop as a collaborative-robot-integrated intelligent welding solution, and the target application will be actual manufacturing sites.
Participating consortium members will divide functions by organization and develop core technologies to implement the intelligent welding system. Doosan Robotics, the lead research and development institution, will oversee the consortium and design the system integration architecture. Doosan Robotics will also be responsible for developing a collaborative robot welding cell, micro-displacement correction AI, and real-time welding machine control technology.
DEEPEX will develop an on-device edge AI device and software development kit (SDK) based on a domestically produced neural processing unit (NPU). DEEPEX will also support the 3-channel vision data pipeline build, robot controller integration, thermal imaging analysis AI model development, verification of environmental durability in welding environments, and reliability certification. SAIGE will be responsible for developing AI models for 3D vision-based workpiece recognition, welding path correction, and weld bead anomaly detection.
The Industry-Academia Cooperation Foundation of Changwon National University will be responsible for developing AI-based welding path generation and physics-based digital twin technology. The four organizations plan to integrate the technologies developed by each institution into a single intelligent welding system.
Demonstrations will then be carried out in stages at Doosan Enerbility nuclear power plant equipment manufacturing sites during the project period. The goal is to complete quality verification for nuclear power plant equipment in 2029. Commercialization is then scheduled to begin from 2030.
The project is being carried out with support from the Ministry of Trade, Industry and Energy and the Korea Evaluation Institute of Industrial Technology (KEIT). Rather than stopping at technology development, the project was designed to extend to on-site manufacturing demonstrations after R&D and to link AI and collaborative robot technologies with manufacturing processes.
Against this backdrop, DEEPEX is making it a policy to secure real-world manufacturing site applications of physical AI that combines ultra-low-power AI semiconductors with robots. A DEEPEX official said the project will serve as an opportunity to create a concrete real-world manufacturing application example of physical AI based on the combination of ultra-low-power AI semiconductors and robots.
The DEEPEX official also said the company aims to contribute to building an AI computing environment that can efficiently process the AI workloads required by manufacturing sites and operate stably based on DEEPEX NPU and on-device AI technologies. This is interpreted as an effort to broaden the base for real-world application in line with the project's goal of linking it to manufacturing-site demonstrations.
A Doosan Robotics official said the company plans to develop an intelligent welding solution that can be used in manufacturing sites by combining Doosan Robotics’ collaborative robot technology with the AI technologies of participating organizations. The official added that the company aims to contribute to solving manufacturing-site problems such as a shortage of skilled workers and to expanding automation and intelligence in welding processes.
Source: TECHWORLD · Park Gyu-chan
Original: https://www.epnc.co.kr/news/articleView.html?idxno=407819
References
This article was produced with the help of an automated content generation algorithm.
Source: TECHWORLD
View originalThis article was summarized and organized by BizCrush based on the original article from TECHWORLD. For exact quotations and full details, please refer to the original article.