ABB, NVIDIA Partner on 'Physical AI'
ABB Robotics and NVIDIA are teaming up to boost industrial automation using NVIDIA's Isaac robotics stack. The goal: bridge the sim-to-real gap, making robot training faster and cheaper by achieving 99% accuracy between virtual and real-world environments. ABB claims this could cut costs by 40% and time-to-market by 50%.
The partnership extends a relationship that began in 2019, when ABB and NVIDIA started collaborating on AI training for robots. Now, the focus is on creating a unified hardware and software platform to accelerate the development and deployment of robotic solutions. NVIDIA's Isaac Sim, a robotics simulation platform, is central to this effort, enabling realistic virtual environments for training robots. ABB is contributing its deep understanding of industrial automation and robot control, aiming to improve the accuracy of simulations and reduce the need for physical testing. This collaboration arrives as manufacturers face increasing pressure to automate and optimize their operations. A recent Deloitte study found that 94% of manufacturing executives believe that AI will be critical to their future success. The combined ABB/NVIDIA tech stack will target applications like robotic assembly, material handling, and autonomous navigation in factories and warehouses. Early results suggest the approach can significantly reduce the time and cost of deploying new robotic systems, potentially accelerating the adoption of automation across industries. Beyond industrial applications, the collaboration could influence other areas of robotics, such as logistics, construction, and healthcare. The ability to train robots in simulation and then transfer those skills to the real world is a key challenge in the field. ABB's Robotics division is headquartered in Västerås, Sweden, and employs over 11,000 people in over 50 countries. NVIDIA, while known for its GPUs, has been making a big push into robotics with its Isaac platform and partnerships with companies like ABB.
