Incorporating Machine Learning with Satellite Data to Support Critical Infrastructure Measurement and Sustainable Development - dissertation

Aug 5, 2026

Upload by Eliana

Artificial Intelligence for good motivates this dissertation’s focus on data-driven methods using machine learning and satellite data to support critical infrastructure measurement and sustainable development. Advances in computation and AI architectures enable efficient extraction of high-dimensional patterns from large-scale heterogeneous datasets. With satellite observations available across diverse temporal-spatial resolutions, AI-based analysis helps explain environmental and socioeconomic dynamics at scale. The research leverages learned knowledge and spatiotemporal characteristics from data-rich areas to improve predictions where observations are limited or absent.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages217

Reading Time9 h 7 min

AccessFree to download

Abstract

Background and motivation

Research approach and contributions

Data-driven methods for measurement and development

Incorporating Machine Learning with Satellite Data to Support Critical Infrastructure Measurement and Sustainable Development | Item Type | Dissertation (Open Access) | | --- | --- | | Authors | Muhebwa , Aggrey | | DOI | 10.7275/36332652 | | Download date | 2026-04-13 11:27:32 | | Link to Item | [https://hdl.handle. net/20.500.14394/19472](https://hdl.handle. net/20.500.14394/19472) | INCORPORATING MACHINE LEARNING WITH SATELLITE DATA TO SUPPORT CRITICAL INFRASTRUCTURE MEASUREMENT AND SUSTAINABLE DEVELOPMENT A Dissertation Presented by AGGREY MUHEBWA Submi!ed to the Graduate School of the University of Massachuse!s Amherst in partial ful"llment of the requirements for the degree of DOCTOR OF PHILOSOPHY February 2024 Electrical & Computer Engineering © Copyright by Aggrey Muhebwa 2024 All Rights Reserved 􀀎􀀙􀀓􀀛􀀒􀀖􀀕􀀘 􀀏􀀘􀀜􀀔􀀗􀀙􀀚􀀔 􀀑􀀎􀀋 􀀊􀀏􀀇􀀏􀀍􀀄􀀅􀀈􀀂􀀃􀀈􀀉􀀅􀀂􀀆􀀏􀀉􀀎􀀂􀀌􀀎􀀇􀀐􀀂􀀄􀀌􀀅􀀐􀀎􀀈􀀏 INCORPORATING MACHINE LEARNING WITH SATELLITE DATA TO SUPPORT CRITICAL INFRASTRUCTURE MEASUREMENT AND SUSTAINABLE DEVELOPMENT A Dissertation Presented by AGGREY MUHEBWA Approved as to style and content by: Jay Taneja, Chair | Colin J. Gleason, Member
| | --- | | Paul Siqueira, Member
| Gabriel Cadamuro, Member Christopher Hollot, Chair of the Faculty Electrical & Computer Engineering DEDICATION To my parents, Alfred and Jane Munyabwera, for se!ing an unparalleled example of excellence. To Billy “Blanks” Kaye (RIP), for opening my eyes to the world of unlimited possibilities through computer programming. And to aunt Merinah Kateeba (RIP) for teaching me the importance of approaching every situation with a positive a!itude. ACKNOWLEDGMENTS “Sometimes people appear in your life unexpectedly like a gi! from the Universe. You didn’t even know you needed them, or that you had called out silently to them. "ey appear when you needed them most, to li! you, educate you, wake you up, or shine a light on your path."ey sprout the seed that was in you, and patiently watch that seed emerge from the soil. Sometimes it wears them out to water and fertilize you every single day as you grow. "isis a delicate time, you as the plant, and they as the nurturer. You as the plant need them foryour growth, and they as your nurturer, have to have the energy to believe in your growth."en, one day you blossom, and awaken to the beauty around you and rejoice. "e only thing you ask from them anymore, is to celebrate the \#ower they have brought to life, and to accept the riches you now will give to them.” -Rii$a Klint. Growing up in rural western Uganda, not in a million years would I have ever imagined that I would one day earn a Ph.D. from one of the world’s top universities. Although I knew from a young age that I wanted to become an engineer and put in the required work, it has been a collective e\#ort of many people that has brought me to this moment. First, I would like to thank my Ph.D. Advisor, Prof. Jay Taneja. $ank you for taking a chance on me and for your patience, kindness, and gentle guidance over the years. You have helped me to become the researcher I am today. To the people who have mentored me through life; Sara Muhahala, Billy Kaye (RIP), and Dr. Je\# Dean, who took me on a%er Billy passed.$ank you for believing in me, even when I didn’t believe in myself. To the members of the STIMA Lab, especially my batchmates: Santiago, June, Zeal, and Bob Muhwezi. $ank you for your support and friendship. To my collaborators; Prof. Taneja, Dr. Cadamuro, Dr. Dongmei Feng, Dr. Lukuyu, and Prof. Colin J. Gleason, thank you for teaching me how to do research correctly. To my commi!ee members, Prof. Taneja, Dr. Cadamuro, Prof. Gleason, and Prof. Paul Siqueira, thank you for your guidance and feedback. Tomy family in Amherst: Dr. Jimi B. and Tolu Oke, Dr. Favorite Iradukunda, Antoine and Angelique Nzeyimana, and the Rwandan Students Association in Amherst area, thankyou for being my family in a foreign land. I am eternally grateful to my parents

FAQs

What is the main purpose of the dissertation?

To develop data-driven machine learning methods using satellite data for critical infrastructure measurement and sustainable development.

How does the work leverage satellite data?

It uses remotely sensed data across varying temporal-spatial resolutions to analyze environmental and socioeconomic changes at scale.

How are predictions improved in data-scarce regions?

The methods transfer knowledge and learned spatiotemporal characteristics from data-rich regions to improve data-driven predictions where data are scarce or unavailable.

Aug 5, 2026

Upload by Eliana