Implementation of smart irrigation using IoT and Artificial Intelligence

Abstract

Water management is crucial for agriculture, as it is the primary source of irrigation for crops. Effective water management can help farmers to improve crop yields, reduce water waste, and increase resilience to drought. This can include practices such as precision irrigation, using sensors and technology to deliver water only where and when it is needed, and conservation tillage, which helps to reduce evaporation and retain moisture in the soil. Additionally, farmers can implement water-saving techniques such as crop selection, crop rotation, and soil conservation to reduce their water use. Thus, studies aimed at saving the use of water in the irrigation process have increased over the years. This research suggests using advanced technologies such as IoT and AI to manage irrigation in a way that maximizes crop yield while minimizing water consumption, in line with Agriculture 4.0 principles. Using sensors in controlled environments, data on plant growth was quickly collected. Thanks to the analysis and training of these data between several models among them, we find the K-Nearest Neighbors (KNN), Support Vector Machine (SVM) and Naive Bayes (NB), the KNN has shown interesting results with 98.4 accuracy rate and 0.016 root mean squared error (RMSE).