Sensor-based Smart Irrigation and IoT Technologies in Precision Agriculture: Current Advances, Challenges and Future Perspectives—A Review
D. Sai Gangadhara Rao *
Acharya NG Ranga Agricultural University, Guntur, AP State, India.
Ch. Murali Krishna
Acharya NG Ranga Agricultural University, Guntur, AP State, India.
E. Manjunatha
Acharya NG Ranga Agricultural University, Guntur, AP State, India.
*Author to whom correspondence should be addressed.
Abstract
Water scarcity, climate variability and increasing competition for freshwater have created a need for more precise irrigation management in agriculture. Conventional irrigation scheduling may not adequately account for spatial and temporal variations in soil moisture, crop water requirements and weather conditions. Sensor-based smart irrigation offers an alternative by combining field measurements, wireless communication, automated control and data-driven decision-making. This review examines advances in soil-moisture and weather-sensing technologies, Internet of Things (IoT) architectures, communication networks, automated irrigation controllers and evapotranspiration-based scheduling. Particular attention is given to the integration of reference evapotranspiration (ET₀), crop evapotranspiration (ETc), crop coefficients and real-time soil-moisture observations to support irrigation decisions. The review also discusses potential applications of artificial intelligence and machine learning models, including artificial neural networks, Random Forest, Gradient Boosting, Extreme Gradient Boosting, LightGBM, Support Vector Regression, Long Short-Term Memory networks, fuzzy logic, and reinforcement learning, for predicting water demand and improving irrigation scheduling. Their advantages, limitations, validation requirements and suitability for different field conditions are considered. The review also examines reported outcomes on irrigation water savings, water-use efficiency, crop productivity, energy consumption, and operational reliability, recognising that results vary by crop, soil, climate, system design, and management practices. Key challenges include sensor calibration, data quality, communication failures, model generalisation, installation costs, interoperability and limited long-term field validation. Future research should prioritise affordable technologies, reliable sensor integration, explainable predictive models and multi-season field evaluation. Integrating agronomic principles with IoT-enabled monitoring and validated AI-based decision support could improve the precision and reliability of open-field irrigation while supporting sustainable agricultural water management.
Keywords: Artificial intelligence, evapotranspiration, internet of things, machine learning, open-field agriculture, precision irrigation, soil-moisture sensors, water-use efficiency