COMING SOON! PQDT Open is getting a new home!

ProQuest Open Access Dissertations & Theses will remain freely available as part of a new and enhanced search experience at

Questions? Please refer to this FAQ.

Dissertation/Thesis Abstract

Considering Methods for Object Reckoning in Vision Equipped Autonomous Flight Systems
by Koutroulis, Anthony, M.S., University of California, Santa Cruz, 2020, 71; 28149695
Abstract (Summary)

The popularity of autonomous flight systems continues to grow as research demonstrates an increasingly broad range of applications with a spectrum of technologies that diminish the need for specialized equipment. Reducing reliance on expensive hardware and decreasing redundancy in sensors is made possible by leveraging the growing compute power of embedded systems and by the adoption of increasingly complex algorithms. Supporting architectures and software systems can increase feasibility by providing safe and robust development environments while increasing system confidence by providing diverse simulation environments for testing flight systems that have been built on new technologies. This work identifies two such categorical technologies, SLAM and machine learning, and surveys contemporary research concerning their use in assisting autonomous flight systems content with foreign objects. Additionally, this work presents a survey of supporting software used in such systems and demonstrates the integration of SLAM and machine learning through simulated environments.

Indexing (document details)
Advisor: Teodorescu, Mircea
Commitee: Elkaim, Gabriel, Dowla, Farid
School: University of California, Santa Cruz
Department: Computer Engineering
School Location: United States -- California
Source: MAI 82/8(E), Masters Abstracts International
Subjects: Computer Engineering
Keywords: Autonomous Flight, Drones, Machine Learning, Object Detection, Quadcopters, SLAM
Publication Number: 28149695
ISBN: 9798569965137
Copyright © 2021 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy