LG Innotek, Nvidia Co-Develop Robot and Car Sensing

Camera modules and sensor boards on a lab bench, with a blurry robot hand in the background

Written by

As of October 22, 2026, LG Innotek is developing visual sensing technology with Nvidia for robots and self-driving cars. The Korean components maker debuted related products at Nvidia's event in Berlin and says it will extend the partnership to parts for humanoid robots.

3 megapixelsResolution of the camera module shown at GTC Berlin

4 companiesOutside partners LG Innotek has signed for physical AI so far (3 if you exclude Aeva)

Key points

LG Innotek showed off new sensor products on Oct. 20 at Nvidia's GTC Berlin tech event in Germany, and made the announcement official on Oct. 22.
The display included a camera module in the 3-megapixel class and a multimodal vision system that combines several sensors to read the surrounding environment.
The two companies plan to widen their work to cover humanoid vision systems and sensing for robot hands, and to unveil products compatible with Nvidia's next-generation AI platform at CES and GTC next year.

At the center of the deal is Nvidia's Holoscan Sensor Bridge (HSB), a data-transfer technology for sensors. Normally, what a camera or sensor picks up passes through the central processor (CPU) before reaching the graphics processor (GPU) that handles AI computing. HSB skips that detour and sends the data straight through. Less waiting in the middle means less time between a robot or self-driving car seeing what's in front of it and deciding what to do.

LG Innotek showed camera modules and a combined vision system built on the technology at the event. The company thinks the payoff will be biggest for robots, which have to react in real time when a person or obstacle suddenly moves. Next, it plans to apply the same approach to the vision devices that serve as a humanoid's eyes and to the sensing tech used in its hands.

LG Innotek built its name on smartphone camera parts and has been pushing that optical design know-how into mobility and robotics. On the self-driving side, it formed a strategic relationship last year with U.S. lidar maker Aeva, and this year it began development work with Applied Intuition and Kakao Mobility. In robotics, it is co-developing components with Boston Dynamics, and it is reportedly also in mass production of camera modules for a North American customer.

CEO Moon Hyuk-soo called the collaboration a chance to introduce the company's differentiated multi-sensor technology to the global market. LG Innotek wants to be a one-stop supplier that covers not just camera sensing but also communications components and semiconductor substrates, and it says it intends to get to the physical AI market first.

In Korea right nowIn Korea, the news is being read as a sign that LG Innotek, long known as a smartphone camera supplier, is shifting its center of gravity toward robots and autonomous driving. With smartphone growth cooling, the big question is whether physical AI (AI that acts in the real world through robots and machines) can become a new profit engine. The Nvidia tie-up is also seen as a way to open doors to overseas robot and vehicle makers.

FAQ

What is HSB technology?

It's an Nvidia communications technology that sends data from cameras and other sensors straight to the AI-computing GPU without going through the CPU. Less lag means faster reactions from robots and self-driving cars.

When will the new products be unveiled?

The two companies plan to show sensors that work with Nvidia's next-generation AI platform at CES and GTC next year.

What kind of company is LG Innotek?

It's an LG Group components maker that built its business on smartphone camera modules. More recently it has been applying multi-sensor technology, including lidar and radar, to robots and autonomous driving.