AGRARIAN develops core technological components towards a dynamic, integrated platform for sustainable agriculture. The project focuses on implementing an accessible and scalable digital ecosystem to support remote and underserved farming communities, while enabling the aggregation of distributed agricultural data for collective policy-making.
One of the primary technical outcomes to date is the development of a user-centric web portal that allows farmers to seamlessly access and deploy applications from a curated repository. This portal is designed with simplicity in mind, lowering the entry barrier for non-technical users and enabling them to select agricultural tools based on their immediate needs (e.g. animal detection, irrigation monitoring or crop health assessment).
In technical level, the platform utilizes a lightweight Kubernetes-based orchestrator (K3s) to dynamically deploy and manage these applications at the network edge. This orchestration layer automates application installation and scaling, adapting to the available edge resources and ensuring efficient service delivery with minimal user intervention.
To address connectivity challenges in rural environments, AGRARIAN integrates hybrid communication technologies, including terrestrial networks (Wi-Fi, 4G/5G) and non-terrestrial links, such as low-earth orbit satellite systems with onboard processing power (e.g. cube satellites). This ensures reliable and redundant communication, even in the event of local network failure, supporting uninterrupted operation of the platform and minimal latency.
The edge processing layer enables real-time analysis of sensor and drone-collected data near the data source. This reduces delays and bandwidth consumption while increasing system responsiveness. A sample use case already validated involved drone-based video streaming for sheep detection, processed in real time via an AI model running on a local edge server.
Finally, AGRARIAN has introduced the Agricultural Decision Support System (ADSS). This platform component consume selected outputs from multiple farm applications, aggregates them across farm clusters and delivers strategic insights optimized for policy makers. These insights support decision-making at regional or national level, moving beyond individual farm optimization to systemic agricultural improvement.