ABB and Nvidia partner to transform factory robot training with physically accurate simulation

ABB teams up with Nvidia to improve factory robot training

ZURICH, March 9 (Reuters) - ABB's (ABBN.S), opens new tab robotics business has partnered with Nvidia (NVDA.O), opens new tab to narrow the gap between how industrial robots perform in virtual simulations and how they behave on factory floors, the companies said on Monday. Swiss-based ABB will use Nvidia's Omniverse libraries of simulated data to make its training environments more realistic by incorporating details such as lighting, shadows, and textures. ABB Robotics President Marc Segura said ⁠robots often have limited information about the world around them, which can undermine accuracy, repeatability and speed. Segura cited the example of a factory robot working close to a stamping machine, which created massive vibrations which reduced the robot's performance. Over time the robot could learn or be programmed how to deal with the vibration, but the technology meant it will already be trained virtually and "know from day one," Segura told Reuters. "This will save companies a lot of time and money." The development is part of a ⁠growing trend where companies run more of their production planning and robot setup in digital simulations to spot problems before equipment starts operating. ABB said the system, delivered via its robot control software, could cut costs and speed time to market by reducing the need for ⁠physical prototypes of products and assembly lines. The technology, due for launch in the second half of 2026, is expected to serve customers in sectors including automotive and consumer electronics. Electronics contract manufacturer ⁠Foxconn (2354.TW), opens new tab is already piloting the technology to install side buttons into consumer electronics, a task ABB said was previously difficult because shadows hindered robot vision. "The industrial sector ⁠needs physically accurate simulation to bridge the gap between virtual training and the real-world deployment of AI-driven robotics at scale," said Deepu Talla, vice president of robotics and edge AI at Nvidia. Reporting by John Revill, editing by Dave Graham Our Standards: The Thomson Reuters Trust Principles., opens new tab

ABB Robotics Taps NVIDIA Omniverse to Deliver Industrial‑Grade Physical AI at Scale

ABB Robotics RobotStudio, enabled by NVIDIA Omniverse libraries, closes the sim‑to‑real gap with 99% accuracy, as Foxconn and global manufacturers begin pilots ahead of its 2026 release. ABB Robotics and NVIDIA today announced a breakthrough partnership that brings industrial‑grade physical AI to the factory floor. By integrating NVIDIA Omniverse libraries directly into its RobotStudio programming and simulation suite, ABB Robotics will now deliver physically accurate simulation capabilities in its platform, dramatically cutting engineering time, reducing deployment costs by up to 40% and accelerating time to market by as much as 50%. The new product -- called RobotStudio HyperReality -- will be available in the second half of 2026 and is already drawing strong interest from ABB Robotic's global customer base. Early pilots include Foxconn, the world's largest electronics manufacturer, and Workr, a U.S.‑based robotic workforce company bringing advanced automation to small and medium-size manufacturers. The partnership marks a major milestone for the industrial sector, which has long sought a reliable way to bring AI-powered intelligence to robots, bridging the sim‑to‑real gap that separates virtual robot training from real‑world performance. "Combining RobotStudio with the physically accurate simulation power of NVIDIA Omniverse libraries, we have closed technology's long-standing 'sim-to-real' gap - a huge milestone to deploying physical AI with industrial-grade precision, for real-world customer applications," said Marc Segura, president of ABB Robotics. A Breakthrough in Physical AI for Industry ABB's integration of NVIDIA Omniverse libraries into RobotStudio brings physically accurate, photorealistic simulation directly into the tool used by more than 60,000 robotics engineers worldwide. The result is a unified workflow where manufacturers can design, program, test and validate entire automation cells before deploying a single robot. RobotStudio HyperReality exports a fully parameterized robot station -- robots, sensors, lighting, kinematics and parts -- as a USD file into NVIDIA Omniverse. There, ABB Robotics' virtual controller runs the same firmware as the physical robot, enabling 99% correlation between simulation and real‑world behavior. Synthetic images generated in Omniverse feed directly into AI training pipelines, allowing vision models to be trained entirely in simulation. This combination of physics‑rich simulation, synthetic data generation and ABB's Absolute Accuracy technology -- which reduces positioning errors from 8-15 mm to around 0.5 mm -- delivers unmatched precision for industrial‑grade applications. Closing the Sim‑to‑Real Gap For decades, manufacturers have struggled with the limitations of simulation: lighting that doesn't match reality, materials that behave differently on the factory floor and models that fail when exposed to real‑world variation. ABB Robotics integration of NVIDIA Omniverse directly addresses these challenges. "The industrial sector needs high‑fidelity simulation to bridge the gap between virtual training and real‑world deployment of AI‑driven robotics at scale," said Deepu Talla, vice president of robotics and edge AI at NVIDIA. "Integrating NVIDIA Omniverse libraries into RobotStudio brings advanced simulation and accelerated computing to ABB's virtual controller technology, accelerating how thousands of manufacturers bring complex products to market." With RobotStudio HyperReality, manufacturers can design and validate production lines virtually, cutting setup and commissioning times by up to 80% and eliminating the need for physical prototypes. The result is faster product ramps, lower cost and greater reliability -- especially for industries like consumer electronics where precision is paramount. ABB Robotics is also exploring the integration of the NVIDIA Jetson edge AI platform into its Omnicore controller to enable real‑time inference across its robot portfolio. Real‑World Pilots: Foxconn and Workr Several customers are already testing RobotStudio HyperReality ahead of its full release. Foxconn is piloting the technology in consumer electronics assembly, where delicate metal components and frequent product variations make automation challenging. Using HyperReality, Foxconn trains robots virtually with synthetic data, achieving unparalleled accuracy when deployed on the production line. The company expects to reduce setup time and eliminate costly physical testing. Workr, a California‑based robotic workforce company, is integrating their own physical AI platform, WorkrCore, with ABB industrial robots trained with synthetic data generated using NVIDIA Omniverse libraries to deploy advanced automation to small and medium-size manufacturers. At NVIDIA GTC 2026 in San Jose, Workr plans to demonstrate AI‑powered robotic systems that can onboard new parts in minutes and deploy without programming expertise. Don't miss NVIDIA founder and CEO Jensen Huang's GTC keynote at the SAP Center on March 16 at 11:00 a.m. PT, where he'll share the latest breakthroughs in AI and accelerated computing. Explore GTC robotics sessions and catch ABB Robotics on the 'Building the Future of Manufacturing' panel as they share how these technologies are shaping the future of intelligent automation.

ABB teams up with Nvidia to improve factory robot training

ABB and Nvidia are joining forces to make industrial robot simulations more realistic. This collaboration aims to bridge the gap between virtual training and actual factory performance. The new technology will help companies save time and money by reducing the need for physical prototypes. A pilot program with Foxconn is already underway. ABB's robotics business has partnered with Nvidia to narrow the gap between how industrial robots perform in virtual simulations and how they behave on factory floors, the companies said ⁠on Monday. Swiss-based ⁠ABB will use Nvidia's Omniverse libraries of simulated data to make its training environments more realistic by incorporating details such as lighting, shadows, and textures. ABB Robotics President Marc Segura said robots often have limited information about the world around them, which can undermine accuracy, repeatability and speed. Segura cited the example of a factory robot working close to a stamping machine, which ⁠created massive vibrations which reduced the robot's performance. Over time the robot could learn or be programmed how to deal with the vibration, but ⁠the technology meant it will already be trained virtually and "know from day one," Segura told Reuters. "This will save companies a lot of time and money." The development is part of a growing trend where companies run more of their production planning and robot setup in digital simulations to spot problems before equipment starts operating. ABB said the system, delivered via its robot control software, could cut costs and speed time to market by reducing the need for physical prototypes of products and assembly lines. The technology, due for launch in the second ⁠half of 2026, is expected to serve customers in sectors including automotive and consumer electronics. Electronics contract manufacturer Foxconn is already piloting the technology to install side buttons into consumer electronics, a task ABB said was previously difficult because shadows hindered robot vision. "The industrial sector needs physically accurate simulation to bridge the gap between virtual training and the real-world deployment of AI-driven robotics at scale," said Deepu Talla, vice president of robotics and edge AI at Nvidia.