In recent years, the proliferation of unmanned aerial vehicles (UAVs) has increased dramatically. UAVs can accomplish complex or dangerous tasks in a reliable and cost-effective way but are still limited by power consumption problems, which pose serious constraints on the flight duration and completion of energy-demanding tasks. The possibility of providing UAVs with advanced decision-making capabilities in an energy-effective way would be extremely beneficial. In this paper, we propose a practical solution to this problem that exploits deep learning on the edge. The developed system integrates an OpenMV microcontroller into a DJI Tello Micro Aerial Vehicle (MAV). The microcontroller hosts a set of machine learning-enabled inference tools that cooperate to control the navigation of the drone and complete a given mission objective. The goal of this approach is to leverage the new opportunistic features of TinyML through OpenMV including offline inference, low latency, energy efficiency, and data security. The approach is successfully validated on a practical application consisting of the onboard detection of people wearing protection masks in a crowded environment.

The global push for green technologies and digital transformation has intensified the demand for critical raw materials, prompting the European Union to enact the Critical Raw Materials Act (CRMA) to ensure ethical, transparent, and resilient mineral supply chains. As the world advances toward the 6G and IoT era, characterized by ultra-low latency, ubiquitous connectivity, and intelligent edge computing, there is an urgent need to empower all levels of the supply chain, especially artisanal small-scale mining (ASM), which remains largely disconnected from such innovations. To address this gap, we propose the Raw Material Radar (RMR) framework, centered on a smart barrel that integrates IoT sensing, edge AI processing, and blockchain-based record-keeping. This low-power device enables real-time detection of shipment anomalies such as tampering or unauthorized access, even in offline and resource-limited environments. Our fully functional prototype achieves high performance (94% F1-score, 98% recall), with minimal resource usage (8.45KB RAM, 407KB flash) and fast inference (5.25 ms), making it well-suited for deployment at the network edge. By enabling verifiable, autonomous custody tracking, the RMR solution exemplifies how edge AI can enhance supply chain transparency and regulatory compliance—critical components for ethical sourcing in the 6G-driven IoT landscape.