In the wake of the Computex 2024 announcements, the partnership between NVIDIA and South Korea’s Doosan Group has shifted from a standard tech collaboration to a comprehensive blueprint for the next industrial era. This expanded alliance targets the convergence of 'Physical AI,' autonomous robotics, and the infrastructure required to power the next generation of 'AI Factories.' By leveraging Doosan’s century-old industrial pedigree and NVIDIA’s accelerated computing dominance, the two entities aim to bridge the gap between digital intelligence and heavy-duty physical labor.
The centerpiece of this collaboration is the integration of the NVIDIA DSX™ platform, which provides the architectural scaffolding for AI factories. Doosan Robotics and Doosan Bobcat are set to integrate NVIDIA’s Isaac Sim and Cosmos world models to refine their autonomous operating systems. This move is designed to evolve industrial robots from pre-programmed machines into intelligent agents capable of sensing, reasoning, and learning in complex, dynamic environments. The scope includes not only traditional factory arms but also emerging forms like dual-arm and humanoid robotic platforms.
Critically, the partnership addresses the 'elephant in the room' for the AI revolution: energy consumption. Doosan Enerbility will play a pivotal role by providing power infrastructure solutions specifically optimized for AI data centers. This includes evaluating the use of Small Modular Reactors (SMRs) and other low-carbon energy sources to sustain the massive electricity demands of NVIDIA’s accelerated computing clusters. The ability to pair high-performance computing with dedicated, reliable power represents a significant strategic advantage in the race to build sustainable AI infrastructure.
Beyond robotics and energy, the collaboration extends to the foundational materials of the hardware itself. Doosan Corporation’s Electro-Materials division will provide advanced Copper Clad Laminates (CCL), a critical component for the high-performance printed circuit boards used in NVIDIA’s AI accelerators and networking equipment. As signal integrity becomes increasingly difficult to maintain at higher bandwidths, Doosan’s specialized materials will serve as the physical bedrock for NVIDIA’s MGX modular server architectures, completing the loop from raw materials to autonomous machines.