Naver's research lab uses AR technology and robot for 3D indoor map data

By Lim Chang-won Posted : August 11, 2020, 09:26 Updated : August 11, 2020, 09:26

[Courtesy of Naver Labs]


SEOUL -- In a crowded underground shopping area, the artificial intelligence wing of South Korea's largest web portal operator has used a self-driving mapping robot, artificial intelligence and augmented reality technology to create 3D maps that would be used for indoor guidance services.

Under a business agreement with the Seoul city government, Naver Labs has tested AR navigation technology at a shopping area in a subway station in Gangnam, a crowded commercial district in southern Seoul. A self-driving robot collected indoor map data and artificial intelligence learned characteristic data such as the edge of pillars to pinpoint exact locations.

Naver Labs has developed an advanced visual localization technology for accurate measurement even in a crowded area. New algorithms that add camera images to LiDAR data were applied to obtain map data on stairs and escalators. LiDAR is a remote sensing technology that measures the distance to a target by illuminating the target with laser light and measuring the reflected light with a sensor.

By applying AR contents to collected 3D map data, Naver Labs would implement a service that makes it easy to find shopping malls, exits and other indoor facilities.

In June, Naver Labs created a 3D layout map for the creation of high-definition digital information on roads, buildings and on-road facilities to help autonomous vehicles navigate in Seoul, which has many narrow streets and maintains the shape of their ancient forms that squirm through residential areas and market places.

Naver Labs has used AI and some 25,000 high-definition aerial photographs to create a 3D model map of the capital city with some 600,000 buildings. High-definition maps would be used for autonomous vehicles or delivery robots.

High-definition digital maps are regarded as a lighthouse that will guide autonomous vehicles around urban areas and help unmanned vehicles communicate through networks to optimize traffic lights or detect pedestrians on the sidewalk after sunset.

Copyright ⓒ Aju Press All rights reserved.