▲ ETRI Level 4 fully autonomous vehicle AutoBee (Photo credit: ETRI)


"Will contribute to the development of the domestic AI and autonomous driving industry ecosystem"
Goal to deploy autonomous driving technology in logistics, public safety, national defense, etc.

While the Electronics and Telecommunications Research Institute (hereinafter ETRI) completed preparations for the era of true Level 4 autonomous vehicles without a driver's seat by unveiling 'Autobee' last year, efforts to showcase more advanced technology are continuing in 2022.

In an interview with this publication, when asked about the goals for 2022, an official from ETRI stated that they are conducting research to ensure the vehicle can drive without problems even on irregular roads (side roads), such as country roads with unclear lanes, and in adverse weather conditions.

It was reported last year that one of ETRI's goals is to enable the dissemination of autonomous driving technology to various fields, such as logistics, public safety, and national defense, through Autobi.

Sensor accuracy is essential for safe driving on irregular roads and in adverse weather conditions, as well as for applications in national defense.

To this end, we have developed core AI technologies that recognize and predict driving environments and situations based on the fusion of cameras and LiDAR, and provide AI learning and inference models to implement perception and prediction functions.

By providing data that can train the model, transformation into various forms It was possible to apply autonomous driving software technology through this.

The Autobi vehicle is equipped with two cameras at the front and two LiDARs each at the front, sides, and rear.

One of its features is that it is equipped with six LiDARs to eliminate blind spots and enhance stability.

The camera sensor detects road surface markings such as lanes and traffic lights on the road, and recognizes whether the taillights of a stationary vehicle are hazard lights or reverse lights.

Lidar is used to recognize dynamic objects (people, vehicles, etc.) in 3D form using artificial intelligence.

An official cited the fact that the Autobee does not travel at high speeds and the limitations of radar in clearly distinguishing between cars and people as reasons for the lack of radar.

Information received from cameras and LiDAR is processed in real-time by AI algorithms applied to AutoBee to recognize the surrounding environment and objects, and autonomously generate a driving path.

According to an official, there is no need to install radar because, under current laws, autonomous vehicles without a driver's seat are limited to a maximum speed of 25 km/h for safety reasons and must maintain this speed even on public roads.

In addition, current laws prohibit fully autonomous driving in child protection zones. It was reported that in the case of Autobi without a driver's seat, it passes through child protection zones by manually switching modes using an in-vehicle joystick (remote control).

AutoBee is also equipped with ETRI's AI-based voice conversation interface technology for user convenience.

The passenger can summon the car to Autobi as if speaking to an AI assistant, or control it by issuing desired commands such as "Let's go to the destination," "Stop," or "Dodge" after getting in.

ETRI equipped a transparent OLED display installed on the window inside Autobi with AR realistic guide technology and 8K VR broadcasting technology.

Passengers can travel without getting bored by receiving real-time vehicle information and content linked to 3D space, or by enjoying 8K high-definition 360-degree VR broadcasts.

The technology developed by the research team is expected to be widely utilized as a platform service that provides real-time, ultra-realistic content while traveling to a destination, even in situations where a car is unavailable or driving is not possible.

For this research, ETRI established 100,000 km of domestic road traffic environment data and produced 200 terabytes (TB) of 14 million training data images.

While sharing relevant data with research institutions and related companies, we continue research on algorithm performance improvement, stabilization, and optimization to contribute to the development of the domestic artificial intelligence and autonomous driving industry ecosystem. It is planned to contribute.

The research is being conducted as part of the Ministry of Science and ICT’s ICT Convergence Industry Innovation Technology Development Project and ETRI’s R&D support project, and is being utilized as a foundation for the Autonomous Driving Technology Development Innovation Project to contribute to achieving the goal of commercializing fully autonomous vehicles by 2027.