Government releases autonomous driving related data for AI learning

The 480 million data for domestic AI learning, which includes 21 kinds of data for autonomous driving such as domestic road driving videos, will be opened from 18th through the AI hub.

The data for artificial intelligence learning provides large-scale data that is even difficult for large companies to obtain on their own, as well as small and medium-sized enterprises and start-ups that are difficult to directly construct data due to the cost and manpower securing issues. It is expected that it will be able to solve the ‘data thirst’, which has been the biggest obstacle in the domestic artificial intelligence industry.

The Ministry of Science and ICT (Minister Lim, Hye-sook) and the Korea Intelligence and Social Promotion Agency (Director Moon, Yong-sik) announced that 170 kinds (480 million cases) of AI learning data, which will be an accelerator for the development of domestic AI technology and industry, will be opened public from June 18 through the artificial intelligence center (aihub.or.kr).

Most of all, the autonomous driving data (21 kinds) which is essential for building smart city mobility services are included.

The autonomous driving data to be released this time is expected to further advance the development of autonomous vehicles by providing various data such as parking obstacles & mobile vehicle recognition images, bus route driving images, and domestic road driving videos.

In particular, the fact that a variety of objects such as special lanes, obstacles, and potholes are included in the data is considered to be the most important advantage.

The 21 types of autonomous driving data including driving videos, objects & obstacle images are large-scale autonomous driving data based on domestic real roads which is the core of the intelligent mobility industry such as autonomous vehicles and drones.

In addition, 19 kinds of safety-related data such as CCTV video, abnormal behavior, traffic flow & disaster situation report will be released. This data is essential to build a proactive and intelligent response system regarding disasters and safety accidents.

Reported by 'Smart City Today'