Detection of UAV hijacking and malfunctions via variations in flight data statistics

Abstract

Detection of potential hijackings of Unmanned Aerial Vehicles (UAVs) is an important capability to have for the safety of the future airspace and prevention of loss of life and property. In this paper, we propose using basic statistical measures as a fingerprint to flight patterns that can be checked against previous flights. We generated baseline flights and then simulated hijacking scenarios to determine the extent of the feasibility of this method. Our results indicated that all of the direct hijacking scenarios were detected, but flights with control instability caused by malicious acts were not detected.