LG Innotek Launches Full-Scale Operation of Autonomous Driving Fleet: "Securing Leadership in Physical AI with Real-World Data"

Direct Operation of Autonomous Driving Data Collection 'Fleet'
Fleet Expanded to 20 Vehicles by 2028 in Partnership with Applied Intuition
Advancing the Camera, Radar, and LiDAR Triad

LG Innotek has teamed up with U.S. autonomous driving software company Applied Intuition to officially launch the operation of its 'autonomous driving sensing data collection vehicle fleet,' which directly collects real-world data on the road.

On September 29, LG Innotek announced plans to gradually expand and operate the fleet, which includes vehicles for data collection and software development, to a total of 20 units by 2028, starting with major roads across the Seoul, Incheon, and Gyeonggi metropolitan area.

LG Innotek's autonomous driving sensing data collection vehicle fleet is running through the city. LG Innotek

This fleet operation aims to advance the performance of sensors—the "eyes" of autonomous vehicles—by utilizing multisensing data collected from real-world driving environments with cameras, radars, and LiDAR.

In particular, LG Innotek is focusing on enhancing the completeness of its proprietary 'sensor fusion' technology. Sensor fusion is a core convergence technology that seamlessly integrates and calibrates the high-precision image information from cameras, distance and speed measurements from radars, and three-dimensional spatial data from LiDAR in real time. This enables accurate recognition of the surrounding environment even in exceptional or unexpected situations.

The company is also concentrating its efforts on securing 'edge case' data, which is considered the biggest challenge in achieving autonomous driving. Edge cases refer to extreme and complex situations—such as black vehicles in heavy rain, strong backlighting at tunnel exits, or damaged lanes in construction zones—where sensor recognition rates drop sharply. Acquiring and analyzing this data is crucial to the reliability of the sensing system. To protect personal privacy, sensitive information such as pedestrians' faces and vehicle license plates, which may be captured during data collection, is immediately de-identified.

LG Innotek plans to combine this accumulated, high-quality real-world data with precision virtual environments (digital twins) to validate sensor systems. Leveraging its top-tier smartphone camera technology, the company is also accelerating its strategy to lead the next-generation physical AI market.

Globally, only a handful of companies possess and supply all three core module technologies—camera, radar, and LiDAR. It is especially rare in the industry for a sensing component solutions company to directly operate a dedicated fleet for data collection. By leveraging the unrivaled sensor fusion competitiveness gained through this fleet operation, LG Innotek aims to accelerate its expansion of orders from global original equipment manufacturers (OEMs) in North America, Europe, and beyond.

Min Jon, Chief Technology Officer (CTO) of LG Innotek, said, "Autonomous driving solutions are evolving beyond simply improving the performance of individual sensors to a new stage of data-driven sensor-software integrated optimization," and emphasized, "We will swiftly lead the next-generation physical AI market with our advanced multi-sensing solutions."

This content was produced with the assistance of AI translation services.