ABB boosts RobotStudio with NVIDIA Omniverse libraries
ABB Robotics is integrating NVIDIA Omniverse libraries into RobotStudio to enhance robot simulation and streamline real-world deployments. The capability will be released as the subscription-based RobotStudio HyperReality in the second half of 2026.
ABB said RobotStudio is used by more than 60,000 robotics engineers worldwide to design robot cells, program robots offline, and simulate production processes. The companies said improved simulation and AI training workflows could make automation more practical for small and mid-sized manufacturers.
With the new integration, robot cells, including robots, sensors, parts, and lighting, can be exported to Omniverse environments for physics-based simulation and rendering. ABB said the improved simulation, combined with its Absolute Accuracy calibration technology, can reduce robot positioning errors to around 0.5 mm in calibrated systems. It added that this helps maintain consistency between simulation and deployed applications.
“RobotStudio HyperReality makes industrial-grade physical AI ready for real-world deployment at scale,” said Marc Segura, president of ABB Robotics. “It can significantly accelerate product development, cutting setup times by up to 80%, reducing costs by 40%, and enabling concurrent engineering without physical prototypes—ultimately speeding time-to-market by around 50% for complex products.”
ABB’s virtual controller runs the same firmware as its physical robot controllers, allowing developers to test motion programs and automation workflows in simulation before deploying them to production systems. ABB claims it is the only robot manufacturer with a virtual controller running the same firmware as the hardware. Segura said this makes the partnership with NVIDIA unique compared to NVIDIA’s deals with other robot arms makers such as Fanuc, Universal Robots, Yaskawa and others.
ABB plans to offer RobotStudio HyperReality as a subscription alongside the existing RobotStudio tiers. The platform currently includes a free version and a paid RobotStudio Premium offering used by industrial customers.
ABB is selling its robotics group to Softbank for $5.375 billion. The deal was announced in October 2025 and is expected to close in mid-to-late 2026. SoftBank is buying ABB’s robotics business to expand its robotics and AI capabilities. Segura said today’s deal with NVIDIA had nothing to do with the pending sale to Softbank.
“[The partnership with NVIDIA] aligns very well with our future strategy, but this development actually stems from a long‑standing collaboration with NVIDIA,” he said. “NVIDIA is powering our autonomous mobile robots for visual SLAM and autonomous navigation, and we’ve been discussing this integration for some time. Two technologies are finally maturing: simulation technology and generative AI. When you bring those together, it makes perfect sense to accelerate.”
Industrial robots have traditionally been deployed in high-volume production environments, where the cost of programming and commissioning systems can be spread across large production runs. In high-mix manufacturing environments, the engineering effort required to program robots often limits adoption. The companies said RobotStudio HyperReality could reduce setup time and engineering effort for new applications.
NVIDIA said ABB is also exploring the integration of the NVIDIA Jetson edge AI platform into its Omnicore controller to enable real‑time inference.
Segura said Foxconn is piloting the technology for consumer electronics assembly, where robots must handle small components and frequent product changes. Another early user is Workr, Segura said, which develops automation systems for small and mid-sized manufacturers.
“Precision is everything in consumer electronics manufacturing and until now, this level of accuracy and fidelity just wasn’t possible in simulation and digital twins,” said Dr. Zhe Shi, chief digital officer of Foxconn. “We’re incredibly excited by the potential of ABB Robotics and NVIDIA’s collaboration, which enables parallel engineering for better designs, faster production ramp‑up and greater product evolution through advanced AI inference and understanding.”
ABB said it is also exploring the use of language and vision-language models inside RobotStudio to simplify robot development. The company said these tools could allow users to instruct robots through prompts rather than traditional programming, with tasks validated in simulation before deployment.
“The goal is to move beyond traditional programming. A proof of that is that within RobotStudio, we’re already using language models so you can prompt robots instead of manually programming them. Our aim is to demonstrate the benefits of AI not only in what robots can do, but also in how people use them,” Segura said. “Over time, the RobotStudio interface will leverage language models so users can move from programming robots to simply instructing them. You simulate the robots, and once training is complete, you push the results to the physical robot. With the new RobotStudio HyperReality product, we’re taking a significant step toward making robots more accessible and easier to deploy.”
And, of course, no robotics press conference is complete without at least one reporter asking the unavoidable question: “When are you building a humanoid?”
“We already have more than 150 different shapes and size of robots. And it’s not just because we like complexity. It’s because industrial customers require optimization. If you want to deploy at scale, everything will have to be optimized. So you can expect from us, of course, mobile manipulation embodiments. Will that be in the shape of a humanoid or a dog? Customers will tell us. It’s going to be a response to their request for optimal solutions.”
The post ABB boosts RobotStudio with NVIDIA Omniverse libraries appeared first on The Robot Report.
