Foxconn pilots ABB–NVIDIA AI robotics platform for electronics assembly
ABB has announced a collaboration with Nvidia to integrate Nvidia Omniverse libraries into ABB’s RobotStudio software platform, aiming to enable large-scale deployment of AI-driven industrial robotics.
The integration will combine ABB Robotics’ programming and simulation environment RobotStudio with the physics-based simulation capabilities of Nvidia Omniverse. The goal is to close the so-called “sim-to-real” gap — the difference between virtual simulations and real-world performance — allowing manufacturers to train robots in digital environments before deploying them on production lines, according to a press release.
The companies say the resulting system, called RobotStudio HyperReality, will enable developers to simulate robotic workflows in digital twins and generate synthetic data for training AI models. ABB states that the technology can achieve simulation accuracy of up to 99% when transferred to physical robot operations.
RobotStudio HyperReality is scheduled for release to ABB’s global RobotStudio user base in the second half of 2026. The company says the technology could reduce production setup and commissioning times by up to 80%, lower costs by as much as 40% and shorten time-to-market for complex products by up to 50%.
The collaboration also includes evaluating integration of the Nvidia Jetson edge computing platform into ABB’s Omnicore robot controllers to enable real-time AI processing at the edge.
EMS provider Foxconn is piloting the first joint application of the technology in consumer electronics assembly. According to ABB, the system allows robots to be trained virtually using synthetic data before deployment, which could reduce setup time and engineering effort when introducing new product variants.
US-based robotics company WORKR is also using the platform to develop robotic manufacturing systems aimed at small and medium-sized manufacturers.
