Back to search

University of Thessaly (UTH)

Optimizing and automating mission execution in drone-based systems

Abstract

dc:description

During the last years, unmanned aerial vehicles (drones), polycopters in particular, are used in an increasing number of civilian applications. This trend is expected to continue thanks to rapidly decreasing acquisition costs, simple deployment, high maneuverability and significant customization potential. Namely, drones can be equipped with various sensors and actuators, they can feature Wi-Fi, 4/5G and custom RF modules for the wireless communication with base-stations and servers that are nearby or in the cloud, and they can carry different embedded platforms to perform computations and data processing onboard, according to the requirements of the application at hand. This thesis focuses on autonomous drones that visit a set of pre-specified locations of interest to perform a sensing task, and, if needed, an additional action (further sensing or some actuation) on the spot. To decide whether such an action is needed, the drones must process the sensed data to detect objects or situations that are relevant for the application at hand. The respective computations can be done locally using an onboard computer, or they can be offloaded to a more powerful server. We study how to minimize the completion time (makespan) of such missions by (i) reducing the flight time of the drones, (ii) reducing the hovering time over the points of interest due to waiting for the computation to complete, and (iii) offloading computations to nearby edge servers to accelerate data processing. These problems become more challenging when considering the limited autonomy of drones and the need to recharge or switch batteries during the mission. Additionally, the shared servers used by drones to offload computations may have limited computational resources thus requiring suitable scheduling for computation offloading in conjunction with path planning. Furthermore, there may be uncertainty regarding the drones’ flight and processing times, introducing changes in the mission plan and server usage. We capture these problems in a formal way, propose solutions and evaluate them through a wide range of simulation experiments, configured using realistic parameters obtained from corresponding software-in-the-loop setups as well as real drone behavior in field tests. Last but not least, we propose a modular software architecture for an autonomous remote sensing system based on drones, which fully automates the entire cycle of operation, including handling of user requests, mission planning, take-off, navigation, sensing, landing, charging and processing the data collected by the drones. Notably, the software can split the mission into parts that can be executed in parallel by multiple drones, with adaptive planning in case a drone experiences an issue and must land earlier than anticipated. The complete software has been tested using software-in-the-loop simulations but also in the field using a real drone which takes-off and lands from/on an automatically controlled hangar.

Degree

thesis:*
Grantor dc:publisher
University of Thessaly (UTH)
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Polychronis, Georgios
  • Πολυχρόνης, Γεώργιος

Subjects

dc:subject × 35

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/58542

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Polychronis, Georgios; Πολυχρόνης, Γεώργιος. Optimizing and automating mission execution in drone-based systems. University of Thessaly (UTH), 2025. http://hdl.handle.net/10442/hedi/58542