Stochastic Model Predictive Control for Irrigation: Addressing Solar and Rain Uncertainties to Enhance Sustainable Productivity

This work addresses a challenging agricultural control problem: to take into account environmental uncertainties (precipitation and solar radiance) in irrigation policies. To tackle these uncertainties, a stochastic model predictive control approach is designed, wherein each type of uncertainty is addressed using two different techniques tailored to effectively counteract them. Simulation experiments were conducted using real-world data spanning various types of days to validate the efficacy of the proposed approach. The results were benchmarked against other methods, showcasing the significant advantages of the proposed approach in terms of accuracy and robustness in agricultural irrigation control in the face of uncertainties. Therefore, this probabilistic approach also offers an effective solution to manage uncertainties and water resources, enhancing the productivity and sustainability of the sector.