Universal Robots has teamed up with NVIDIA to add new AI capabilities, making it easier for customers to deploy collaborative robots for low-volume, high-mix production.

Universal Robots, the world's No. 1 collaborative robot company, announced on the 19th (local time) at the NVIDIA GTC 2024 (GPU Technology Conference), the world's largest AI conference held in San Jose, California, USA, that its parent company Teradyne is collaborating with NVIDIA, the world's largest AI semiconductor company with a 90% share of the AI semiconductor market, to add new AI functions to collaborative robots and mobile robots.

Teradyne, the parent company of Universal Robots, is an automation company headquartered in North Reading, Massachusetts, USA, and its main industries are robots and semiconductors. In the semiconductor sector, it has global semiconductor companies such as Samsung Electronics, Qualcomm, Intel, and IBM as its customers, and in the robot sector, it owns Universal Robots and Mir, an autonomous mobile robot company.

After three years of collaboration with NVIDIA, Universal Robots is now able to support programming that combines NVIDIA’s AI technology. This will allow customers to easily introduce collaborative robots into low-volume, high-mix production by simplifying the setup of common industrial applications.

Collaborative robots can also improve inspection for manufacturers of automobiles, large electronics, and home appliances. Teradyne and Nvidia cited ease of programming, reduced computational time for trajectory planning, optimization, and execution as benefits of combining AI semiconductors and collaborative robots.

Universal Robots has integrated NVIDIA accelerated computing into its collaborative robots (cobots) to enable path operations that are 50 to 80 times faster than current applications, and will demonstrate an autonomous inspection system using cobots and AI at GTC 2024. In the GTC 2024 demo, a camera mounted on a Universal Robots cobot was shown inspecting a workpiece that changed direction randomly, while a digital twin mirrored the cobot’s movements.

The collaboration between the two companies combines the NVIDIA Jetson AGX Orin onboard Apollo; the cuMotion path planner with NVIDIA’s Isaac Manipulator, a state-of-the-art robotic arm perception, path planning and motion control library; PolyScope X software, and the UR5e cobot platform.

Universal Robots users can load CAD files for up to 20 parts along with associated test procedures when using inspection applications. The company said the goal was to “improve existing collaborative robot programming methods by introducing NVIDIA solutions and enable a variety of applications that were previously impossible to fully automate.”

“This collaboration marks a milestone in the convergence of collaborative robotics and AI,” said Ujjwal Kumar, CEO of Teradyne Robotics. “We are continuing to invest in Universal Robots to establish it as the preferred robotics platform for developing and deploying AI applications by adding high-performance computing hardware to our control systems and investing in targeted upgrades to our software stack.”

He continued, “We have always sought to transform our customers’ lives in the ways that matter most to them. Creating platforms for new solutions to problems that were previously unsolvable is part of that effort. Teradyne Robotics, a leader in industrial automation, will combine with NVIDIA’s cutting-edge AI semiconductor technology to design a new future for robotics.”