AI-enabled hydroponics: a critical review of sensor-driven farming systems for sustainable global food security.

Abstract

Main conclusion: This review concludes that integrating AI with smart sensing has the potential to substantially enhance hydroponic production through precise monitoring, adaptive control, and improved resource-use efficiency. Hydroponics, a key approach in controlled-environment agriculture (CEA), enables precise regulation of plant growth conditions, improving resource efficiency and year-round crop production. Recent advances in artificial intelligence (AI) and sensor technologies offer opportunities to enhance nutrient management, monitor plant health, and predict physiological responses in real time. This review critically evaluates AI-enabled hydroponic systems, focusing on sensor-driven monitoring of solution chemistry (EC, pH, DO), environmental parameters, and plant-health indicators through imaging. We highlight the application of time-series forecasting models, computer vision, and reinforcement learning for adaptive control, emphasizing their impact on plant growth, nutrient uptake, and stress mitigation. Key challenges, including sensor drift, model generalization across crops, and adoption barriers, are discussed. Finally, we outline strategies for integrating AI with plant-centered hydroponic design, including hybrid control frameworks, prototype concepts for explainable AI dashboards, and open datasets. The dashboard concept is presented as a future design direction to improve transparency and operator trust rather than as a validated user interface evaluated in this review. The accompanying conceptual implementation example illustrates how the reviewed AI methodologies could be integrated within an AIoT-enabled hydroponic system, while quantitative validation across multiple crops, cultivation cycles, and production environments remains necessary before generalizing their effectiveness.