ABB Robotics is integrating NVIDIA Omniverse libraries into RobotStudio to reinforce robotic simulation and streamline real-world deployments. The aptitude might be launched because the subscription-based RobotStudio HyperReality within the second half of 2026.
ABB stated RobotStudio is utilized by greater than 60,000 robotics engineers worldwide to design robotic cells, program robots offline, and simulate manufacturing processes. The businesses stated improved simulation and AI coaching workflows may make automation extra sensible for small and mid-sized producers.
With the brand new integration, robotic cells, together with robots, sensors, elements, and lighting, may be exported to Omniverse environments for physics-based simulation and rendering. ABB stated the improved simulation, mixed with its Absolute Accuracy calibration know-how, can cut back robotic positioning errors to round 0.5 mm in calibrated methods. It added that this helps keep consistency between simulation and deployed functions.
âRobotStudio HyperReality makes industrial-grade bodily AI prepared for real-world deployment at scale,” stated Marc Segura, president of ABB Robotics. “It could considerably speed up product improvement, slicing setup occasions by as much as 80%, decreasing prices by 40%, and enabling concurrent engineering with out bodily prototypesâfinally rushing time-to-market by round 50% for advanced merchandise.”
ABBâs digital controller runs the identical firmware as its bodily robotic controllers, permitting builders to check movement packages and automation workflows in simulation earlier than deploying them to manufacturing methods. ABB claims it’s the solely robotic producer with a digital controller working the identical firmware because the {hardware}. Segura stated this makes the partnership with NVIDIA distinctive in comparison with NVIDIA’s offers with different robotic arms makers comparable to Fanuc, Common Robots, Yaskawa and others.
ABB plans to supply RobotStudio HyperReality as a subscription alongside the prevailing RobotStudio tiers. The platform at the moment features a free model and a paid RobotStudio Premium providing utilized by industrial prospects.
ABB is promoting its robotics group to Softbank for $5.375 billion. The deal was introduced in October 2025 and is anticipated to shut in mid-to-late 2026. SoftBank is shopping for ABBâs robotics enterprise to broaden its robotics and AI capabilities. Segura stated as we speak’s cope with NVIDIA had nothing to do with the pending sale to Softbank.
â[The partnership with NVIDIA] aligns very effectively with our future technique, however this improvement truly stems from an extendedâstanding collaboration with NVIDIA,” he stated. “NVIDIA is powering our autonomous cell robots for visible SLAM and autonomous navigation, and weâve been discussing this integration for a while. Two applied sciences are lastly maturing: simulation know-how and generative AI. Once you deliver these collectively, it makes good sense to speed up.”
Industrial robots have historically been deployed in high-volume manufacturing environments, the place the price of programming and commissioning methods may be unfold throughout massive manufacturing runs. In high-mix manufacturing environments, the engineering effort required to program robots usually limits adoption. The businesses stated RobotStudio HyperReality may cut back setup time and engineering effort for brand new functions.
NVIDIA stated ABB can also be exploring the combination of the NVIDIA Jetson edge AI platform into its Omnicore controller to allow actualâtime inference.
Segura stated Foxconn is piloting the know-how for client electronics meeting, the place robots should deal with small parts and frequent product modifications. One other early consumer is Workr, Segura stated, which develops automation methods for small and mid-sized producers.
“Precision is every thing in client electronics manufacturing and till now, this stage of accuracy and constancy simply wasnât doable in simulation and digital twins,” stated Dr. Zhe Shi, chief digital officer of Foxconn. “We’re extremely excited by the potential of ABB Robotics and NVIDIAâs collaboration, which permits parallel engineering for higher designs, sooner manufacturing rampâup and higher product evolution via superior AI inference and understanding.”
ABB stated it is usually exploring the usage of language and vision-language fashions inside RobotStudio to simplify robotic improvement. The corporate stated these instruments may enable customers to instruct robots via prompts slightly than conventional programming, with duties validated in simulation earlier than deployment.
“The aim is to maneuver past conventional programming. A proof of that’s that inside RobotStudio, weâre already utilizing language fashions so you’ll be able to immediate robots as an alternative of manually programming them. Our goal is to display the advantages of AI not solely in what robots can do, but in addition in how folks use them,” Segura stated. “Over time, the RobotStudio interface will leverage language fashions so customers can transfer from programming robots to easily instructing them. You simulate the robots, and as soon as coaching is full, you push the outcomes to the bodily robotic. With the brand new RobotStudio HyperReality product, weâre taking a major step towards making robots extra accessible and simpler to deploy.”
And, in fact, no robotics press convention is full with out no less than one reporter asking the unavoidable query: “When are you constructing a humanoid?”
“We have already got greater than 150 totally different shapes and measurement of robots. And it’s not simply because we like complexity. It’s as a result of industrial prospects require optimization. If you wish to deploy at scale, every thing must be optimized. So you’ll be able to count on from us, in fact, cell manipulation embodiments. Will that be within the form of a humanoid or a canine? Prospects will inform us. It’s going to be a response to their request for optimum options.”
The publish ABB boosts RobotStudio with NVIDIA Omniverse libraries appeared first on The Robotic Report.
