Blockchain based UAV MarketPlace with Machine Learning for Livestock Health Prediction

Food safety and public health are fundamentally intertwined with livestock health and welfare. Consequently, continuous monitoring and data-driven assessment of animal health status are critical to mitigating disease risks and enabling proactive, well-informed decision-making among relevant stakeholders. Yet, current digital livestock solutions remain largely fragmented, with sensing, data management, and analytics components developed independently. However, this paper presents DroneVet, a comprehensive solution that integrates heterogeneous technological building blocks orchestrating : 1) data acquisition through IoT devices and UAVs, 2) secure and trusted data storage and management via a blockchain-based marketplace, and 3) an analytics component that assesses livestock health status based on drone collected data. We implement a proof-of-concept using blockchain Ethereum framework and a decentralized application (DApp). We deploy and compare different machine learning algorithms using a real-world dairy cattle dataset from INRAE institute. The results demonstrate the feasibility of combining such heterogeneous technologies for a smart agriculture scenario.