LG Electronics is expanding its robotics collaboration with NVIDIA as it develops a Data Factory in Seoul to support physical AI, robot learning, and commercialization of its robotics business.

LG Electronics announced it is accelerating its robotics collaboration with NVIDIA following a visit by senior NVIDIA officials at its Data Factory, currently under construction at the company’s Yangjae R&D Campus in Seoul.

The visit followed an August 13 memorandum of understanding signed by LG Group and NVIDIA at NVIDIA’s headquarters in Santa Clara, California. LG said the companies reconvened at the Data Factory four days later to review the direction of their collaboration and discuss ways to generate further synergies.

Data Factory to Support Robot Learning

LG officials and NVIDIA representatives reviewed the status of the Data Factory, which is scheduled to be fully operational by the end of the year. LG has deployed its self-developed LG CLOiD home robots at scale to generate, collect, and learn from data.

The facility includes training spaces where robots practice tasks and dedicated areas for validating and refining datasets. It includes a replicated home environment where LG CLOiD learns and repeats cleaning tasks and a simulated manufacturing space modeled on LG’s washing machine plant in Tennessee, where LG CLOiD units move, stack, and assemble parts.

Robots are also being deployed in LG CNS’s logistics automation solutions and in a space where LG Innotek trains robotic hands. LG said data collected across these environments will be linked with NVIDIA’s robotics stack, where the data will be augmented and synthesized into high-quality data for robot learning.

Physical AI and Robotics Commercialization

LG said the core competitiveness of physical AI lies in building a “data flywheel,” where high-quality data is repeatedly secured and then used to train and improve robots. The company expects that combining data from its manufacturing and logistics operations with NVIDIA’s robotics stack will help create a world-class data flywheel.

The Data Factory uses NVIDIA physical AI technologies, including NVIDIA Omniverse libraries, NVIDIA Cosmos world foundation models, and the open NVIDIA Isaac robotics development platform, across the data-to-deployment process. The Yangjae facility spans four floors, including one basement level and three above ground, with a total floor area of 10,000 square meters, and is expected to house several hundred robots by the end of the year.

By the end of the year, LG expects training data collected directly at the facility and data synthetically generated and augmented using NVIDIA Cosmos open world models to total 100,000 hours, equivalent to roughly 12 years of data. LG plans to use the data to advance its Robot Foundation Model, which underpins the performance of humanoid robots.