Date of Award

3-2025

Document Type

Thesis

Degree Name

Master of Science in Electrical Engineering

Department

Department of Electrical and Computer Engineering

First Advisor

David Woodburn, PhD

Abstract

Autonomous aircraft must land without human intervention, but existing methods rely on GPS or marked runways, which may be unavailable in austere environments. This paper presents a vision-based approach using semantic segmentation to detect runways and estimate aircraft pose by comparing camera and satellite imagery. We detail the model’s training and demonstrate its effectiveness with simulated and real UAV data.

AFIT Designator

AFIT-ENG-MS-25-M-034

Comments

An embargo was observed for posting this thesis.

This work is marked Distribution A, Approved for Public Release. PA case number 88ABW-2025-0369

Download

Share

COinS