[Edaily Reporter Park Jung-Soo ] #K3I. Co., Ltd. announced on the 29th that it is moving full steam ahead with the implementation of a physical AI manufacturing infrastructure based on its proprietary platform, “Neuro Twin X,” through an international joint research project being conducted with Siemens, the Korea Electronics Technology Institute (KETI), and others.
Last year, the company was selected as a joint research partner for a government-funded project led by KETI to develop AI-based digital twin technology. Participants in the project include Siemens, a European industrial automation company, as well as German research institutions and companies.
K3I. Co., Ltd. has completed the simulation design for the first year and will begin building a Physical AI manufacturing infrastructure this year, which will allow for the placement and verification of manufacturing site components in a virtual space.
The core of this infrastructure is to replicate key elements of the manufacturing site—such as workers, robots, and manufacturing equipment—in a virtual space, enabling the verification of process flows and equipment operations before actual equipment is installed. The company plans to enhance the design efficiency and operational stability of smart manufacturing sites based on its proprietary digital twin technology.
K3I. Co., Ltd. is building its digital twin implementation capabilities based on high-precision 3D spatial information technology. The company already has a track record of recreating lithium-ion battery manufacturing processes in a virtual space using CAD drawings and actual measurement data, and has improved the accuracy of its digital twins by faithfully reproducing even equipment layout and worker movement patterns similar to those in real-world environments.
The company is also developing a platform that integrates factory digital twins and robot simulations, centered on “Neuro Twin X,” which interfaces with NVIDIA Omniverse. The company explains that the precisely rendered virtual factory allows for the deployment and simulation of various robots under conditions identical to the real environment, making it suitable for use as a basis for physical AI training.
A company official stated, “The second year of the research marks the stage of materializing the simulation design framework into a prototype for physical AI manufacturing infrastructure,” adding, “We are advancing smart manufacturing digital twin technology centered on ‘Neuro Twin X,’ which enables the deployment and verification of workers, manufacturing equipment, and robots in a virtual space.”
The spokesperson continued, “We have verified actual robot control and virtual factory simulations on our proprietary Omniverse integration platform and have secured technology for standardizing physical AI data,” adding, “Through this government-funded project, we will further advance a digital twin manufacturing simulation environment applicable to manufacturing sites and expand our penetration of the physical AI manufacturing infrastructure market.”