How ABB and NVidia Omniverse fix digital twin blind spots

Application Engineering Manager Mark Ruth demonstrates photorealistic HyperReality in RobotStudio bin-picking
Related To: 
Aug. 20, 2026
Add Us On Google

Key Highlights

  • By integrating NVIDIA Omniverse with RobotStudio, engineers can accurately model complex environmental variables like dynamic lighting, reflections and surface textures to train machine vision systems entirely offline.
  • Because RobotStudio and ABB's physical controllers share the same operating system, motion paths and camera triggers developed in hyper-realistic simulation translate directly to real-world hardware without unexpected behavior.
  • Validating complex vision-guided applications, such as 3D bin-picking with cobots, in a digital environment allows integrators to bypass lengthy physical trial-and-error, dramatically cutting deployment times on the factory floor.

At the 2026 Automate show, Mark Ruth, application engineering manager at ABB Robotics, showcased how physical automated systems can be mirrored and trained in virtual environments before reaching the factory floor (Figure 1). His demonstration highlighted the evolution of ABB's RobotStudio software, a tool long used for precise trajectory planning and cycle-time estimation, by introducing its integration with high-fidelity, real-time simulation technology.

Figure 1: Mark Ruth, application engineering manager, ABB Robotics.

Figure 1: Mark Ruth, application engineering manager, ABB Robotics.

Digital twin software has struggled to train machine vision applications because virtual environments lacked accurate surface textures, ambient reflections and dynamic lighting conditions, explains Ruth. By leveraging the NVidia Omniverse platform, ABB’s HyperReality tech preview bridges this gap. Engineers can export RobotStudio stations into photorealistic environments, simulating real-world physics and lighting to train vision software completely offline.

The live physical demonstration includes a GoFa cobot executing a bin-picking application. Fitted with an OnRobot gripper and a Photoneo 3D vision sensor, the real-world robot seamlessly replicated the camera triggers, part detection and motion paths developed inside the digital environment. This workflow proves how hyper-realistic simulation allows integrators to deploy complex vision-guided robotic tasks faster and with confidence.

About the Author

Mike Bacidore
Email

Mike Bacidore

Editor in Chief

Mike Bacidore is chief editor of Control Design and has been an integral part of the Endeavor Business Media editorial team since 2007. Previously, he was editorial director at Hughes Communications and a portfolio manager of the human resources and labor law areas at Wolters Kluwer. Bacidore holds a BA from the University of Illinois and an MBA from Lake Forest Graduate School of Management. He is an award-winning columnist, earning multiple regional and national awards from the American Society of Business Publication Editors. He may be reached at [email protected] 

Sign up for our eNewsletters
Get the latest news and updates

Related

Simulation vs. reality: Can AI and virtual testing replace old-school engineers?
Human in the loop: Teaching tomorrow's engineers to leverage and audit AI design tools
White Paper: How OEMs Can Benefit from Peak Shaving
Sponsored
How IEC IP20 can help protect your employees
Sponsored