Global road map tracks four years of infrastructure change using AI
The United Nations considers well-developed and safe roads an important part of infrastructure in its Sustainable Development Goals. Until now, however, there has been no benchmark for the condition and state of road networks worldwide. Researchers at the Institute of Geography at Heidelberg University and HeiGIT (Heidelberg Institute of Geoinformation Technology) have addressed this gap by using…
Researchers at the Institute of Geography in Heidelberg and HeiGIT have utilized artificial intelligence and satellite imagery to develop an open-access dataset mapping over 9 million kilometers of global roads, capturing changes in road conditions from 2020 to 2024. This dataset, based on high-resolution PlanetScope satellite images, aids humanitarian applications and socioeconomic development assessments, particularly in data-scarce regions.
Using deep learning, researchers identified and classified approximately 9.2 million kilometers of major roads worldwide, with a 20 percentage point improvement in accuracy compared to previous datasets. The model also tracks infrastructure changes over time, enabling assessments of passability under varying conditions, such as extreme weather events.
The data is converted into a Humanitarian Passability Matrix, supporting humanitarian mission planning and equitable urban development. The dataset, available via the United Nations Office for the Coordination of Humanitarian Affairs, can serve as a dynamic measure of global and local socioeconomic progress, highlighting infrastructure gaps and aiding investment decisions for economic development.
Written by urgent.news from Phys.org's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
