Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Design of an Autonomous UAV Tracking System

Tollebäck, Scott and Persson, Vilhelm (2026)
Department of Automatic Control
Abstract
Drones are becoming more common everywhere, and with that comes the risk of unauthorized drones entering protected areas such as airports or military sites. One way to deal with this is to send out another drone to find and follow the unknown one for closer inspection.
This project built a system that does exactly that. A drone equipped with a camera is able to automatically spot another drone, estimate how far away it is, and follow it, all without a human controlling it. The system uses an object detection model to find the drone in the camera image and a tracking method to keep following it between detections.
Tests showed that the system could spot a drone from up to around 80 meters away when the sky was in the background, and... (More)
Drones are becoming more common everywhere, and with that comes the risk of unauthorized drones entering protected areas such as airports or military sites. One way to deal with this is to send out another drone to find and follow the unknown one for closer inspection.
This project built a system that does exactly that. A drone equipped with a camera is able to automatically spot another drone, estimate how far away it is, and follow it, all without a human controlling it. The system uses an object detection model to find the drone in the camera image and a tracking method to keep following it between detections.
Tests showed that the system could spot a drone from up to around 80 meters away when the sky was in the background, and around 20 meters when the background was rocky terrain. Once close enough, the system was able to follow the target drone at a distance of around 10 meters in real time.
The results show that this kind of autonomous drone detection and following is possible using standard hardware and commercially available components, though improvements to the camera pipeline and more diverse training data would make the system more robust in the future. (Less)
Please use this url to cite or link to this publication:
author
Tollebäck, Scott and Persson, Vilhelm
supervisor
organization
year
type
H3 - Professional qualifications (4 Years - )
subject
report number
TFRT-6314
other publication id
0280-5316
language
English
id
9246937
date added to LUP
2026-08-25 10:32:48
date last changed
2026-08-25 10:32:48
@misc{9246937,
  abstract     = {{Drones are becoming more common everywhere, and with that comes the risk of unauthorized drones entering protected areas such as airports or military sites. One way to deal with this is to send out another drone to find and follow the unknown one for closer inspection.
 This project built a system that does exactly that. A drone equipped with a camera is able to automatically spot another drone, estimate how far away it is, and follow it, all without a human controlling it. The system uses an object detection model to find the drone in the camera image and a tracking method to keep following it between detections.
 Tests showed that the system could spot a drone from up to around 80 meters away when the sky was in the background, and around 20 meters when the background was rocky terrain. Once close enough, the system was able to follow the target drone at a distance of around 10 meters in real time.
 The results show that this kind of autonomous drone detection and following is possible using standard hardware and commercially available components, though improvements to the camera pipeline and more diverse training data would make the system more robust in the future.}},
  author       = {{Tollebäck, Scott and Persson, Vilhelm}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Design of an Autonomous UAV Tracking System}},
  year         = {{2026}},
}