LG Innotek to Directly Verify Autonomous Driving 'Eyes' ... Builds Domestic Fleet of 20 Vehicles

Backed by its partnership with Applied Intuition, LG Innotek is ramping up fleet operations in South Korea to directly collect autonomous driving sensing data on public roads. On the 29th, LG Innotek announced the operation of vehicles for collecting domestic autonomous driving sensing data with Applied Intuition. Applied Intuition is a U.S. autonomous driving software company.

The two companies signed a strategic partnership in March to advance autonomous driving sensing solutions. Based on that partnership, LG Innotek has begun operating data collection vehicles centered on Seoul, Incheon and Gyeonggi.

The purpose of the fleet operation is to improve sensor performance based on measured data from real-world road environments. Those sensors serve as the 'eyes' of autonomous vehicles.

LG Innotek plans to expand its domestic fleet to about 20 vehicles by 2028, including data collection vehicles and autonomous driving software development vehicles among future additions.

Global autonomous driving companies are investing heavily in fleet operations. LG Innotek said it is rare for a sensing solutions company to operate its own fleet for data collection.

LG Innotek said it equips its data collection vehicles with sensors such as cameras, radar and LiDAR, and can analyze the data obtained while driving to identify information about road conditions, vehicle movement, object locations and weather changes. The company plans to combine this collected data with Digital Twin.

LG Innotek plans to use the combined data to improve the performance of individual sensors and verify systems. A key focus in this process is advancing sensor fusion technology. Sensor fusion combines and calibrates camera image information, radar distance and speed information, and LiDAR's precise 3D spatial information. As autonomous driving advances, the importance of technology that accurately recognizes the surrounding environment through the fusion of heterogeneous sensor data also grows.

As one of the key goals for achieving fully autonomous driving, LG Innotek is also pushing ahead with securing Edge Case data. Edge cases refer to situations where there are no problems in normal environments, but sensor recognition performance deteriorates under specific conditions.

Autonomous driving sensors can be vulnerable in exceptional situations such as black vehicles in heavy rain, strong backlighting at tunnel exits or irregular lane markings at construction sites. LG Innotek plans to secure data on these various exceptional scenarios, and said the data it secures will be used to improve sensor accuracy and reliability.

LG Innotek is also responding to privacy protection issues during the data collection process. Personal information such as pedestrians' faces and vehicle license plates captured by vehicle cameras is anonymized before use.

LG Innotek has secured real-world driving data through its fleet operations and is moving to target the autonomous driving sensing market based on that data. The target product range is a multi-sensing solution covering automotive cameras, radar and LiDAR. LG Innotek said it can supply all three sensing modules, camera, radar and LiDAR, with its own technology, and that such companies are rare even in the global market. The company plans to strengthen differentiated competitiveness by advancing sensor fusion technology based on actual fleet data it has secured, and accelerate orders from new customers in North America and Europe.

The Applied Intuition data collection vehicles are equipped with LG Innotek sensing modules. These data collection vehicles are scheduled to operate in South Korea, the U.S., Europe and Japan. LG Innotek plans to analyze large-scale measured data accumulated in each region. The goal of the analysis is to preemptively identify areas for improvement across diverse driving environments. LG Innotek plans to reflect the analysis results in sensor module performance.

Min Jeon, CTO of LG Innotek, said the direction of evolution in autonomous driving solutions is shifting from focusing on improving the performance of individual sensors to optimizing sensors and software simultaneously based on data. He also said the company aims to take the lead in the physical AI sensing market, including autonomous driving, with data-driven advanced multi-sensing solutions.

LG Innotek has set autonomous driving and robotics, among other physical AI technologies, as a new growth engine. The company is expanding its physical AI-related businesses. It also holds technological capabilities secured in the smartphone camera field. Based on those capabilities, LG Innotek is expanding its business scope into the market for high-performance sensing solutions for autonomous vehicles and robotic environment recognition.

LG Innotek is broadening its cooperation with external companies to strengthen its autonomous driving capabilities. Last year, it signed a strategic partnership with Aeva, a U.S. LiDAR technology company, and this year it collaborated with Applied Intuition, a U.S. autonomous driving software company, as well as domestic mobility platform company Kakao Mobility. LG Innotek is maintaining that collaboration trend by pursuing joint development of autonomous driving solutions with Applied Intuition and Kakao Mobility.

Source: TECHWORLD · Park Gyu-chan
Original: https://www.epnc.co.kr/news/articleView.html?idxno=407451

References

• https://www.epnc.co.kr/news/articleView.html?idxno=407451