LG, Nvidia to jointly develop humanoid robot for 2027 unveiling

LG will unveil Nvidia-powered bipedal humanoids by the first quarter of 2027 as part of a larger deal to collaborate across multiple AI realms, including AI factories and autonomous driving technologies. 

The humanoid robots will run on Nvidia’s Jetson Thor onboard computing platform and be built using Isaac GR00T, the company’s open humanoid foundation model, and Halos for Robotics, its safety system for robotics.

LG will supply actuators through LG Electronics, sensors through LG Innotek and batteries through LG Energy Solution.

LG and Nvidia signed a memorandum of understanding for business cooperation on Thursday at Nvidia’s headquarters in Santa Clara, California, with LG Group Chairman Koo Kwang-mo, Nvidia CEO Jensen Huang and other senior executives from both companies in attendance. The agreement builds on a meeting between the companies’ top executives at LG’s Seoul headquarters in June.

The two will also operate a joint task force made up of technology and business experts, who will work closely across research and development, field trials and business commercialization.

LG also plans to deploy a wheel-based version of its CLOiD robot on a washing machine production line at LG Electronics’ plant in Tennessee this year to test the technology in a manufacturing environment, then expand its use to other production sites, homes and commercial spaces.

Beyond robotics, LG and Nvidia will build AI factories using Nvidia’s DSX architecture combined with LG’s cooling, battery, power and design capabilities across its affiliates. The companies plan to build a pilot facility in the first half of 2027 based on Nvidia’s next-generation Vera Rubin platform, followed by an 80-megawatt facility in Cheonan, South Chungcheong, in the first half of 2028. The Cheonan site will use prefabricated construction methods that LG said will cut building time by more than 20 percent.

In mobility, LG will combine Nvidia’s Drive Hyperion platform with its in-vehicle infotainment and software capabilities to develop a computing platform for AI-defined vehicles, extending its automotive electronics business into autonomous driving.

“As our collaboration has picked up speed, the tasks that we will tackle together in AI factories, physical AI and mobility have become clear,” Koo said in a statement. “We will accelerate the spread of AI by building references that lead the industry.”

BY LEE JAE-LIM [lee.jaelim@joongang.co.kr]