Recipient organisationUlster UniversitySource-published name: University of Ulster
Funding£32K
PeriodMar 2025 — Jul 2025
In plain English
AI plain-English summary
A drone delivering a defibrillator to a heart attack victim could be hijacked mid-flight, and this project builds a security system to stop that. As drones take on life-critical roles—rushing medical supplies, searching disaster zones—they become tempting targets for hackers who could spoof signals, steal data, or take control. Current cybersecurity tools, designed for static networks, fail to protect fast-moving, autonomous swarms. This project fills that gap by creating a multi-tier, AI-driven framework that watches every layer of a drone operation: the drone itself, its communications with other drones and ground stations, and the cloud system that manages the mission. Machine learning algorithms continuously analyse behaviour, flagging anything unusual—a sudden change in flight path, an unauthorised command—and respond in real time to block the threat. If successful, the system could make drone-based emergency logistics genuinely trustworthy. Paramedics could rely on drone-delivered supplies without worrying about interception. Search-and-rescue teams could coordinate autonomous aircraft without fear of sabotage. The impact is on the quiet infrastructure of emergency response—the invisible chain of data, commands, and trust that keeps a drone on its life-saving course.
View original technical description
The proposed project will deliver an AI-driven, multi-tier cybersecurity framework designed for autonomous drones operating in critical sectors such as disaster relief, search and rescue, and medical deliveries. As drones play an increasingly crucial role in these high-risk operations, they become targets for malicious attacks that exploit vulnerabilities through hijacking, spoofing, and data breaches, potentially jeopardising mission success and public safety. The proposed system will ensure trustworthiness and security at multiple levels within a drone network, continuously evaluating and verifying the integrity of individual drones, securing drone-to-drone and drone-to-infrastructure communication, and safeguarding cloud-based mission control systems. AI-driven anomaly detection and behavioural analysis will be at the core of the framework, enabling real-time identification and mitigation of security threats to maintain operational integrity. Unlike traditional single-tier security solutions, which are often inadequate for highly mobile, autonomous systems, this multi-tier intrusion detection and trust management approach will offer robust protection against evolving cyber risks. The framework will integrate machine learning techniques to detect unauthorised activity, ensuring resilient and tamper-proof drone operations, particularly in challenging environments where reliability is paramount. With the increasing role of drones in emergency healthcare logistics, such as delivering defibrillators and critical medical supplies, it is imperative to guarantee secure and reliable operations. The proposed system will provide next-generation cybersecurity protection, reinforcing trust, safety, and operational resilience in the rapidly evolving field of autonomous aerial systems.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know