The United Nations Institute for Training and Research (UNITAR) provides innovative learning solutions to individuals, organizations and institutions to enhance global decision-making and support country-level action for shaping a better future.
Description of Activities on AI
Project 1: NVIDIA
The United Nations Satellite Centre (UNOSAT) announced a collaboration with NVIDIA on training and research activities to promote the use of Artificial Intelligence (AI) for Earth Observation activities in support of the Sustainable Development Goals (SDGs), with an initial emphasis on disaster management. This cooperation framework allows UNOSAT and NVIDIA to benefit from their respective facilities, resources, and domain experience. The collaboration has two initial priorities: 1) integration of NVIDIA’s accelerated computing platform within UNOSAT’s infrastructure to fast‐track research and development of AI for Earth Observation and 2) design and roll‐out of an e‐learning course on the use of deep learning for flood detection to upskill data scientists within disaster management agencies worldwide.
Project 2: UNOSAT S-1 FloodAI
The UNITAR-UNOSAT designed, developed, and deployed UNOSAT S-1 FloodAI: an end-to-end pipeline where Copernicus Sentinel-1 Synthetic Aperture Radar (SAR) imagery of flood-prone areas are automatically downloaded and processed by a deep learning model to output flood vector data and update operational dashboards. Access to timely and accurate data could not only inform the decision-making process to help optimize the disaster response, but it also has the potential to significantly reduce the loss of life and mitigate structural damage, particularly in the context of humanitarian operations, thus supporting both national authorities and international emergency management organizations for the benefit of affected populations.
Project 3: Mapping Refugee Settlement and Damage Assessment with Machine Learning and Remote-Sensing Data
The purpose of this project is the creation of an end-to-end pipeline that takes high-resolution satellite imagery as input and returns damage assessment maps in the form of a building footprint together with a damage class label.
Project 4: ML4Floods
UNITAR-UNOSAT partner with Trillium Technologies and FDL Europe to test and use ML4Floods: an ecosystem of data, models and code pipelines to tackle flooding with machine learning ML. After a successful testing phase of the methodology, UNOSAT is implementing ML4Floods into its operations and deploying the tool into the UNOSAT AI pipeline at CERN.