# NVIDIA and Doosan say robotics is becoming infrastructure, not just automation hardware Source: TechNewsList (https://technewslist.com) Canonical URL: https://technewslist.com/en/article/nvidia-doosan-physical-ai-factory-stack-2026-06-12-morning Section: Drones & Robots (https://technewslist.com/en/drones-robotics) Author: TechNewsList Language: en Published: 2026-06-13T04:12:37.441+00:00 Updated: 2026-06-13T04:12:37.59337+00:00 > NVIDIA's June 7 Doosan expansion and its early-June physical AI research push suggest robotics is moving beyond standalone machines into a broader infrastructure layer that spans simulation, world models, industrial equipment, energy, and AI factory deployment. ## TL;DR - NVIDIA said on June 7, 2026 that it is expanding collaboration with Doosan Group across robotics, AI factory power systems, and electronics materials for next-generation data centers. - Doosan Robotics plans to combine Isaac Sim, Isaac Lab, Cosmos, Newton, and Jetson Thor in its Agentic Robot OS, while other Doosan units explore physical-AI uses in equipment and energy systems. - That points to a robotics market where value comes from infrastructure-scale stacks, not only from selling individual robot arms or autonomous machines. ## Key points - NVIDIA is extending physical AI from research tools into industrial deployment partnerships. - Doosan is using one collaboration to connect collaborative robots, construction equipment, power systems, and AI factory materials. - Robotics value is shifting toward simulation-to-real pipelines, reasoning systems, and energy-aware deployment rather than isolated device specs. - Industrial players increasingly want reference stacks they can standardize across multiple machine classes. - Physical AI may become a data-center-and-field infrastructure story as much as a robotics product story. # NVIDIA and Doosan say robotics is becoming infrastructure, not just automation hardware ## What happened NVIDIA said on June 7, 2026 that it is expanding collaboration with Doosan Group across physical AI, robotics, AI factory power solutions, and electronics materials for next-generation data center systems. The partnership spans multiple Doosan businesses, including Doosan Robotics, Doosan Bobcat, Doosan Enerbility, and Doosan Corporation Electro-Materials BG. ![Contextual editorial image for NVIDIA and Doosan say robotics is becoming infrastructure, not just automation hardware NVIDIA Doosan Group Doosan Robotics Doosan Bobcat Isaac Sim NVIDIA Blog NVIDIA Blog technology news](https://wimg.heraldcorp.com/news/cms/2026/04/29/news-p.v1.20260429.fd0b87f37e6b4031a5bcc43feaaa9a91_P1.jpg) *Contextual visual selected for this TechPulse story.* The robotics part is the most immediately important. NVIDIA said Doosan Robotics is integrating Isaac Sim, Isaac Lab, Cosmos world models, the Newton physics engine, and Jetson Thor into its Agentic Robot OS. The goal is to create a platform that connects perception, reasoning, simulation, learning, and on-device inference for industrial robots operating in more dynamic real-world environments. But the announcement goes further than robots alone. NVIDIA and Doosan are positioning physical AI as a cross-layer system that also touches construction equipment, AI factory power infrastructure, and the printed-circuit-board materials needed for high-performance data center systems. That widens the story from machine autonomy to industrial infrastructure. ## Why it matters This matters because it shows how the robotics market is changing shape. For years, many companies sold automation hardware as if the machine itself were the product. Physical AI shifts the center of gravity. The real value starts to come from simulation tools, world models, deployable software stacks, edge inference hardware, and the energy and compute systems needed to keep all of it running reliably. NVIDIA understands that and is trying to become the platform layer underneath it. Doosan is a useful partner because it operates across several parts of the industrial world that are normally discussed separately: collaborative robots, compact heavy equipment, energy systems, and advanced industrial materials. One partnership can therefore become a testbed for what an infrastructure-scale robotics strategy looks like. That matters for buyers because it makes physical AI easier to standardize. If simulation, reasoning, deployment, and hardware can be reused across several machine classes, the economics of robotics adoption improve substantially. ## Technical details NVIDIA said Doosan Robotics will use Isaac Sim and Isaac Lab as core frameworks, alongside Cosmos and the Newton physics engine, to improve how robots perceive, reason, and learn in dynamic settings. The company highlighted use cases such as depalletizing and sanding, plus future robot form factors including dual-arm and humanoid systems. ![Contextual editorial image for NVIDIA and Doosan say robotics is becoming infrastructure, not just automation hardware NVIDIA Doosan Group Doosan Robotics Doosan Bobcat Isaac Sim NVIDIA Blog NVIDIA Blog technology news](https://blogs.nvidia.com/wp-content/uploads/2025/06/franka.jpg) *Contextual visual selected for this TechPulse story.* Doosan Bobcat is exploring how the same physical AI stack can support construction, landscaping, agriculture, and material handling equipment. That suggests a shared world-model approach rather than siloed autonomy for each machine type. The power and data-center side is also notable. Doosan Enerbility is exploring support for NVIDIA AI factories through gas turbines, steam turbines, small modular reactors, and fuel-cell systems. Meanwhile, Doosan's electro-materials business is supporting the MGX ecosystem with copper clad laminate for printed circuit boards used in AI accelerators, networking, and AI server motherboards. Taken together, those pieces show an unusual level of vertical integration. Simulation and autonomy software sit at one end, and power delivery plus data-center materials sit at the other. That is exactly what makes the announcement look like infrastructure rather than another robot demo. ## Market / industry impact The broader market implication is that industrial AI vendors may increasingly compete as systems integrators of intelligence, energy, and compute. A company that can provide only a robot, only a chip, or only a simulator may not capture the highest-value part of the stack once customers want scalable deployment. This also pushes robotics closer to the economics of cloud infrastructure. Physical AI systems need reliable power, high-speed electronics, scalable simulation, and consistent deployment workflows. The companies that can align those layers may define the next industrial platform winners. For traditional robotics vendors, that raises the bar. Buyers will expect more than mechanical performance. They will want adaptable reasoning, simulation-backed deployment, and a clear path to operating fleets in changing environments without rewriting everything from scratch. ## What to watch next Watch whether Doosan turns Agentic Robot OS into repeatable production deployments rather than a framework announcement. Real industrial rollouts will matter more than architecture claims. Also watch whether the AI factory and energy side of the partnership deepens. If it does, NVIDIA's physical AI story will look more like industrial infrastructure strategy than a robotics marketing theme. Finally, watch whether other heavy-industry groups build similar alliances with AI platform vendors. If they do, the next robotics cycle will be defined by stack partnerships, not isolated machine launches. ## Sources - NVIDIA Blog, "NVIDIA and Doosan Group Collaborate to Advance Physical AI and AI Factory Infrastructure," published June 7, 2026. - NVIDIA Blog, "NVIDIA Enables the Next Era Of Physical AI Research With Agent Skills For Autonomous Vehicles, Robotics And Vision AI," published June 3, 2026. Mentions: NVIDIA, Doosan Group, Doosan Robotics, Doosan Bobcat, Isaac Sim, Jetson Thor, physical AI ## Sources - [NVIDIA Blog](https://blogs.nvidia.com/blog/nvidia-and-doosan-group-physical-ai/) - [NVIDIA Blog](https://blogs.nvidia.com/blog/cvpr-physical-ai-research-agent-skills/)