0
LG Innotek's autonomous driving sensing data collection vehicle runs on a downtown street. / LG Innotek

LG Innotek announced on the 29th that it will deploy autonomous driving sensing data collection vehicles in South Korea in partnership with U.S.-based autonomous driving software firm Applied Intuition.

The two companies signed a strategic partnership in March to enhance autonomous driving sensing solutions. LG Innotek will initiate operations focused on Seoul, Incheon, and Gyeonggi Province, with plans to operate a fleet of around 20 vehicles—including data collection vehicles and autonomous driving software development platforms—across the nation by 2028.

LG Innotek's direct fleet operation represents a proactive move to upgrade the performance of camera, radar, and lidar sensors, which act as the eyes of autonomous vehicles, based on data collected under real-world driving conditions.

While global autonomous driving companies invest heavily in fleet operations, it is rare across the industry for an autonomous driving sensing solutions provider to directly run its own data collection fleet.

LG Innotek's data collection vehicles are equipped with an array of sensors, including cameras, radar, and lidar. Processing the data collected during road travel yields extensive information recognized by the sensors, such as road conditions, vehicle movements, object positions, and weather shifts. LG Innotek plans to combine the collected real-world data with virtual simulation environments to refine individual sensor performance and validate entire systems.

The company is focusing on advancing its proprietary sensor fusion technology. Sensor fusion combines and calibrates image data from cameras, distance and velocity data from radar, and high-precision 3D spatial data from lidar to accurately recognize surrounding environments. As autonomous driving capabilities mature, fusing and integrating data across disparate sensors becomes just as crucial as individual sensor specifications.

At the same time, LG Innotek plans to focus on securing edge-case data essential for full autonomy. Edge cases refer to scenarios where sensing performance degrades under specific conditions despite functioning properly in normal environments. Notable examples include black vehicles operating in torrential downpours, severe glare at tunnel exits, and irregular lane markings in construction zones. Such data plays a pivotal role in boosting sensor accuracy and reliability across diverse anomaly situations.

LG Innotek is also taking thorough precautions regarding personal data protection. Personal information captured by vehicle cameras during data collection, such as pedestrian faces and vehicle license plates, is de-identified prior to use.

Leveraging the data obtained through fleet operations, LG Innotek plans to accelerate its penetration into the autonomous driving sensing market. Armed with comprehensive sensing solutions encompassing vehicle cameras, radar, and lidar, the company aims to expedite winning new contracts from clients across North America, Europe, and other global markets.

LG Innotek is among a select few companies worldwide capable of supplying all three sensor modules through proprietary in-house technologies. The company intends to widen the competitive gap further by coupling these capabilities with advanced sensor fusion technology fueled by real-world fleet driving data.

"Autonomous driving solutions are evolving from merely advancing individual sensor performance toward co-optimizing sensors and software based on data," said John Min, CTO and Senior Vice President at LG Innotek. "By leveraging our advanced, proprietary multi-sensing solutions built on real-world data, we will swiftly capture the physical AI sensing market, including autonomous driving."

                                                                                                             Ahn So-yeon