Anti-Drone and Anti-Autonomy: Achieving Drone Control via System Logic Analysis

Drones (more formally known as unmanned aerial vehicles or unmanned aerial systems) are poised to provide benefits in numerous areas of society. Their application areas range from package delivery to military surveillance to scientific data collection. However, drone presence and activities are, in some cases, undesirable. Whether launched by a prankster, criminal, terrorist organization or state actor – or perhaps just an average citizen who has lost control of the craft – there is a need to prevent drone operations. This paper proposes a method that does not require physical interaction with the drone or the ability to compromise its ground control station or its onboard security or other software. Instead, the proposed method focuses on the identification of the logical processes that are used for decision making, their decomposition into rules (including rules that must be represented using partial membership and fuzzy logic) and the mapping of the rules into an expert system-style network. From this network, a solver algorithm can be utilized to identify solutions that modify external inputs (electronic data and sensed information) to produce a desired response from the UAV's autonomous command system. These responses could include departure from a restricted area, positioning for in-air capture or targeting, landing or a targeted crash. This paper presents the proposed system, the logic-to-rule decomposition process and a solver mechanism for the rule-based system. In also discusses the implementation of the system and its testing, before concluding with a discussion of its efficacy for various applications and pathways for future work.