University of North Texas
Control-Aware Manipulation of ArduPilot via Legitimate MAVLink Commands: Simulation and Hardware Validation
Abstract
dc:descriptionThis thesis reports about investigating control-aware attacks against the ArduPilot UAV control system, where an attacker exploits flight-controller dynamics using only legitimate MAVLink commands. Unlike prior work centered on malformed packets, channel compromise, or physical sensor attacks, our approach targets trusted run-time interfaces such as parameter updates, estimator configuration, and guidance commands. We develop six attacks that manipulate oscillatory behavior, Proportional-Integral-Derivative (PID) gains, Extended Kalman Filter (EKF) operation, failsafe assumptions, yaw control, and multi-layer controller interactions. Using Software-In-The-Loop (SITL) simulation and Pixhawk 2.4.8 hardware validation, we define the control-aware attack model for ArduPilot, design and evaluate six attacks, validate them in simulation and hardware experimentation, and identify security weaknesses in parameter handling and command trust. Thus, we show that malicious well-formed MAVLink messages can cause measurable degradation in attitude stability, angular-rate behavior, trajectory tracking, and in severe cases can drive the vehicle into unsafe states.
Degree
thesis:*- Grantor dc:publisher
- University of North Texas
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Benchellal, Feras
- Contributors dc:contributor
-
- Tymoshchuk, Pavlo
- Morozov, Kirill
- Othmane, Lotfi Ben
- Tunc, Cihan
Subjects
dc:subject × 3Rights
dc:rights- Statement dc:rights
-
- Public
- Benchellal, Feras
- Copyright
- Copyright is held by the author, unless otherwise noted. All rights Reserved.
- Language dc:language
- English
Identifiers
dc:identifier.*- Identifier
-
local-cont-no: submission_4985
ark: ark:/67531/metadc2652490