Key points are not available for this paper at this time.
The growing need for food and water throughout the world has put precision agriculture in the spotlight lately. In order to fulfil this demand, farmers will require access to water and fertile land. Farmers need a solution that alters their operations since both resources are scarce. The key to producing larger, better, and more lucrative crops with less resources is precision irrigation. To enhance water utilization efficiency, several machine learning-driven irrigation models have been proposed. However, these models face challenges in regions with unpredictable climates due to their limited learning capabilities. This study introduces an intelligent irrigation system for precision agriculture, leveraging the Internet of Things (IoT) and deep learning neural networks. This system, incorporating feedback mechanisms, exhibits sustained performance over time, independent of local weather conditions. A long short-term memory network (LSTM) is employed to predict irrigation schedules, required water amounts for agricultural land, and the geographic allocation of water. Simulation results unequivocally demonstrate that, in the experimental agricultural area, this model surpasses the efficiency of the most recent models in water utilization.
Sarjerao et al. (Tue,) studied this question.