IOGRUCloud: A Scalable AI-Driven IoT Platform for Climate Control in Controlled Environment Agriculture
The scalable IoT system reduced manual calibration by 73% while managing 2.3M daily sensor events.
Researcher Andrii Vakhnovskyi has published a paper detailing IOGRUCloud, a scalable AI-driven IoT platform designed for precise climate control in Controlled Environment Agriculture (CEA). The system features a three-tier architecture that separates field-level sensing/actuation (L1), facility-level coordination (L2), and cloud-level optimization (L3-L4), enabling a progression from rule-based to fully autonomous operation. Its core innovation is a Vapor Pressure Deficit (VPD) cascading control loop that uses GRU-enhanced PID tuning, which reportedly reduces the manual calibration effort by a significant 73%.
Deployed across 14 production greenhouses totaling 47,000 square meters, the platform demonstrated substantial real-world benefits. It achieved a 23% reduction in energy consumption and a 31% improvement in climate stability compared to baseline systems. The platform proved its scalability by handling 2.3 million daily sensor events while maintaining 99.7% uptime. Vakhnovskyi has released the full architecture specification and deployment results to support reproducibility and further development in the smart agriculture research community, highlighting a move toward more open and efficient AI solutions for sustainable farming.
- Uses GRU-enhanced PID tuning for VPD control, reducing manual calibration effort by 73%
- Deployed across 14 greenhouses (47,000 m²), cutting energy use 23% and improving climate stability 31%
- Scalable architecture handles 2.3M daily sensor events with 99.7% operational uptime
Why It Matters
Demonstrates how AI and IoT can significantly boost efficiency and sustainability in resource-intensive agriculture.