Development of a Distributed Learning and Scalable Testbed for UAVs Using Machine Learning

Jawad, Mahmood (2026) Development of a Distributed Learning and Scalable Testbed for UAVs Using Machine Learning. Doctoral thesis, Dundalk Institute of Technology.

PDF (Jawad Mahmood PhD Thesis) - Accepted Version Download (56MB)

Abstract

This project aims to develop a testbed for designing and training control algorithms for cooperative and self-organizing swarms of UAVs. The primary purpose of developing a distributed learning and scalable testbed based on Machine Learning (ML) algorithms is to enable UAVs to make decisions using real-time data and perform tasks autonomously. In this project, a novel testbed is developed that allows the integration of different ML algorithms with a flight simulator. This testbed supports multiple UAVs that learn to fly and coordinate in a simulated environment to accomplish the objective of target tracking. It employs novel techniques that enable faster learning and higher performance compared to conventional ML methods. FlightGear is the flight simulator used in this project. Once trained, the aircraft can fly with the help of an ML model and show stable behaviour compared to an untrained model. This testbed can be used to train control models for a wide variety of use cases. As proof of concept, a problem is formulated regarding target tracking of UAVs. The tracking aircraft follows the path of the target aircraft. Both tracking and target aircraft are controlled by different ML models and fly on a common flight simulator. This testbed can also scale up the number of tracking aircraft

Item Type:	Thesis (Doctoral)
Subjects:	Computer Science > Computer Software
Research Centres:	UNSPECIFIED
Depositing User:	John Loane
Date Deposited:	01 Sep 2026 08:12
Last Modified:	01 Sep 2026 08:12
License:	Creative Commons: Attribution-Noncommercial-Share Alike 4.0
URI:	https://eprints.dkit.ie/id/eprint/1101

Actions (login required)

View Item

Downloads