Creating Artificial Worlds with AI to Improve Energy Access Data (2021-2022)

Energy infrastructure mapping is vital for well-informed policy decisions around energy access expansion, but data on existing infrastructure can be scarce. Previous work has shown that these information gaps can be filled by applying machine learning techniques to satellite and aerial imagery. However, computer vision techniques struggle when applied across satellite imagery from diverse geographies. More scalable and versatile techniques are needed to assess infrastructure relevant for sustainable energy transitions.

This team set out to develop a better way to detect and map multiple energy infrastructures through publicly available satellite imagery globally by applying synthetic data generation techniques, including generative models and style transfer, for creating synthetic overhead imagery; comparing these techniques with traditional, high-effort techniques for training data collection and recent 3D modeling approaches; and sharing these tools and generated data in an open-source repository to encourage use by other researchers and decision makers.

Ultimately, the team found that their proposed solution using synthetic data generation significantly improves energy infrastructure detection when collection of real data is cost-prohibitive. They hope that their technique will aid efforts to expand sustainable energy.

Learn more about this project team by viewing the team's video.


Fall 2021 – Spring 2022

Team Outputs

Team website

Energy Infrastructure Mapping Using AI (2022 Fortin Foundation Bass Connections Virtual Showcase)

Domain Adaptable Deep Learning Models for Energy Infrastructure Detection using Synthetic Data Generation (poster by Madeleine Jones, Caleb Kornfein, Alex Kumar, Aya Lahlou, Jaden Long, Madeline Rubin, Caroline Tang, Frankie Willard, Alena Zhang, Saksham Jain, Kyle Bradbury, Jordan Malof and Simiao Ren, presented at Fortin Foundation Bass Connections Showcase, Duke University, April 13, 2022)

Self-Supervised Encoders Are Better Transfer Learners in Remote Sensing Applications. Zachary D. Calhoun, Saad Lahrichi, Simiao Ren, Jordan M. Malof, Kyle Bradbury. 2022. Remote Sensing 14, no. 21: 5500.

This Team in the News

Two Faculty Recognized for Exceptional Co-Leadership of Bass Connections Team

See related teams, Tracking Climate Change with Satellites and Artificial Intelligence (2022-2023) and Deep Learning for Rare Energy Infrastructure in Satellite Imagery (2020-2021), and Data+ summer project, Tracking Climate Change Impacts with Satellites and AI (2022).


Image: Wind farm, by pfly, licensed under CC BY 2.0

Wind farm.

Team Leaders

  • Kyle Bradbury, Pratt School of Engineering-Electrical & Computer Engineering|Energy Initiative
  • Jordan Malof, Pratt School of Engineering-Electrical & Computer Engineering

/graduate Team Members

  • Saksham Jain, Electrical/Computer Engg-MS
  • Katherine Wu, Master of Engineering Mgmt-MEG
  • Boya Sun, Interdisciplinary Data Science - Masters
  • Ben Ren, Electrical/Computer Engg-PHD

/undergraduate Team Members

  • Yuxi Long, Mathematics (BS)
  • Yucheng Zhang, Computer Science (BS)
  • Frank Willard, Computer Science (BS)
  • Caroline Tang, Mathematics (BS)
  • Madeline Rubin, Interdisciplinary Progrm (BSE)
  • Aya Lahlou, DKU Interdisciplinary Studies (BS)
  • Alexander Kumar, Computer Science (BS)
  • Caleb Kornfein, Computer Science (BS)
  • Madeleine Jones, Computer Science (BS)

/yfaculty/staff Team Members

  • Leslie Collins, Pratt School of Engineering-Electrical & Computer Engineering
  • Robyn Meeks, Sanford School of Public Policy
  • Luana Lima, Nicholas School of the Environment-Environmental Sciences and Policy
  • Marc Jeuland, Sanford School of Public Policy
  • T. Robert Fetter, Nicholas Institute for Environmental Policy Solutions