Graduation Semester and Year
Summer 2026
Language
English
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Computer Engineering
Department
Computer Science and Engineering
First Advisor
Mohammad Atiqul Islam
Second Advisor
Phuc VP Nguyen
Third Advisor
Nick Gans
Fourth Advisor
Dave Levine
Abstract
Persistent unmanned aerial vehicle operation is limited by the endurance, maintenance, and mass costs of conventional batteries. Solar harvesting offers an alternative, but removing the battery turns flight into a coupled systems problem: guidance decisions simultaneously affect aerodynamic state, mission progress, harvested power, stored energy, and, in multi-aircraft missions, formation behavior. This dissertation develops a research trajectory for energy-aware battery-free solar unmanned aerial systems, progressing from the technical horizon, through a physical aircraft, to repeatable swarm-scale evaluation. First, it identifies the capabilities and research challenges required for UAVs that sustain operation through harvested energy rather than battery replacement or recharging. Second, it presents a battery-free, solar-powered fixed-wing UAV constructed from off-the-shelf components. Mechanical and electrical design-space analysis, together with greedy and predictive energy-aware control, enables the prototype to manage power intermittency and perform flight behaviors using solar energy. Bench-top and outdoor evaluations establish the feasibility of the approach and expose the constraints that make larger-scale physical experimentation difficult. Third, the dissertation presents Swarm CAELUS, a telemetry-first simulation framework for coordinated solar UAV missions. The framework co-evolves guidance, vehicle dynamics, weather exposure, energy storage, formation state, and failure logic while producing synchronized pervehicle evidence. A 32-run matched simulation study shows that a reserve policy can prevent iii early leader power failure and extend operation without restoring mission completion, exposing a trade between electrical survival and flight performance. It also shows that prescribed per-vehicle environmental forcing can increase electrical heterogeneity and change follower-failure ordering even when leader terminal classes remain unchanged. Together, these studies show that battery-free solar UAV autonomy is not an isolated powersystem problem. It requires joint reasoning about energy, airframe, environment, control, and coordination, supported by physical prototypes and reproducible preprototype evaluation.
Keywords
battery-free, energy-aware, UAV, supercapacitors, solar harvesting
Disciplines
Hardware Systems
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Liller, Jackson G., "ENERGY-AWARE BATTERY-FREE SOLAR UNMANNED AERIAL SYSTEMS: FROM SINGLE-AIRCRAFT FEASIBILITY TO SWARM-SCALE EVALUATION" (2026). Computer Science and Engineering Dissertations. 23.
https://mavmatrix.uta.edu/cse_dissertations2/23
Comments
I would like to thank Phuc (VP) Nguyen and Mohammad Atiqul Islam for their mentorship, guidance, and sustained support throughout this research. Their insight helped shape both the individual studies and the larger argument of this dissertation. I am grateful to the faculty and staff who administer The University of Texas at Arlington’s Graduate Assistance in Areas of National Need (GAANN) Fellowship program, and to the U.S. Department of Education support that made the fellowship possible. I also thank the faculty, staff, and students of the Department of Computer Science and Engineering at The University of Texas at Arlington for providing the community, resources, and opportunities in which this work developed