ABB Robotics partners with NVIDIA to bridge the gap between virtual simulation and real-world industrial applications

IT House, March 10 — ABB’s robotics division has partnered with NVIDIA to reduce the gap between virtual simulation performance and actual factory operation of industrial robots.

According to IT House, the Swiss-based ABB will use NVIDIA’s Omniverse simulation database to incorporate details such as lighting, shadows, and textures, making the robot training environment more closely resemble real-world scenarios.

ABB Robotics President Mark Seegula stated that robots typically have limited perception of their surroundings, which can affect their accuracy, repeatability, and speed.

Seegula gave an example: if a robot operates near stamping equipment, the intense vibrations from stamping can reduce its performance. He told Reuters that traditionally, robots had to learn gradually or be programmed to adapt to vibrations, but this new technology allows robots to be trained in a virtual environment, “equipped with the necessary capabilities from day one of deployment,” which will save companies significant time and costs.

A rising industry trend is that more companies are using digital simulations for production planning and robot deployment, allowing issues to be identified before equipment goes into actual operation. This partnership exemplifies that trend.

ABB stated that the system will be delivered through its robot control software, reducing the need for physical prototypes of products and production lines, thereby lowering costs and speeding up time-to-market.

The technology is scheduled to be launched in the second half of 2026, targeting clients in automotive, consumer electronics, and other industries.

Electronics contract manufacturer Foxconn has already piloted the technology for installing side buttons on consumer electronic devices. ABB noted that previously, visual recognition was difficult for such tasks due to shadow interference.

NVIDIA’s Vice President of Robotics and Edge AI, Deep Talla, said, “The industrial sector needs high-precision simulation technology to bridge the gap between AI-driven virtual robot training and large-scale real-world deployment.”