ABB and NVIDIA Aim to Close the Gap Between Robot Simulation and Real Deployment

• ABB Robotics is integrating NVIDIA Omniverse libraries into its RobotStudio platform to train industrial robots in simulated environments before deployment.
• The collaboration introduces RobotStudio HyperReality, a system designed to improve production setup speed and reduce manufacturing costs through virtual testing.

ABB Robotics announced a partnership with NVIDIA to integrate Omniverse libraries into ABB’s RobotStudio platform. The collaboration is intended to help manufacturers train robots in simulated environments before deploying them in real-world robotics applications. By combining ABB’s software, programming, design, and simulation tools with NVIDIA’s accelerated computing and simulation technologies, the companies said the collaboration will help manufacturers deploy physical AI in robotics applications.

The integration will support a new capability called RobotStudio HyperReality. The system uses physically accurate simulation, digital twins, and synthetic data to train robots before deployment in real environments. ABB said the approach can reduce setup and commissioning times by up to 80%, lower costs by up to 40% by reducing the need for physical prototypes, accelerate time-to-market by up to 50%, and help close the sim-to-real gap with up to 99% accuracy.

“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. “Integrating NVIDIA Omniverse libraries into RobotStudio brings advanced simulation and accelerated computing to ABB Robotics’ unique virtual controller technology, accelerating how manufacturers of all sizes bring complex products to market,” said Marc Segura, President of ABB Robotics, in an official news release.

Foxconn is piloting the system for consumer electronics assembly. The company is training assembly robots virtually using synthetic data to refine production processes across different device variants before moving them to the production line with high accuracy. Robotics company WORKR is also using the technology for systems built on ABB robotics, trained with synthetic data using NVIDIA Omniverse libraries and deployed without operators needing to know programming.

🌀 Tom’s Take:

Robots trained in simulation often struggle in the real world because small differences in lighting, materials, and environments can break what worked virtually. Closing that gap lets robots learn tasks in software and move straight into production, reducing costly testing on physical factory lines.

Source: ABB