Intrusion Detection Systems-Enabled Power Electronics for Unmanned Aerial Vehicles

Abstract

Compromised power electronics, due to firmware attacks and hardware Trojans, in a flight computer can jeopardize the safety and security of an Unmanned Aerial Vehicle (UAV). They can maliciously alter sensor measurements or control commands to make a UAV to take disastrous moves. Unfortunately, most of these attacks are difficult to detect before deploying components in the system. Therefore, detecting compromised behavior run-time is important, while it is challenging at the same time. In this work, we propose to build machine learning-based intrusion detection systems (IDSs) to be deployed at the power electronics/microcontorller level such that it can deal with malicious data/control commands initiated due to hardware attacks.