Architecture for customizable irrigation systems using Machine Learning
irrigation; IoT; machine learning.
The global industry has been going through successive transformations due to constant technological developments and innovations in its processes. As a result, technological advances are increasingly accepting existing processes and it is common to seek the development of techniques that help in the automation of procedures that were previously done entirely manually. This occurs in agriculture, which plays an important role in the economic development of a country like Brazil. However, rural activities are quite unpredictable, due to soil dynamics or climatic variations, where events contribute to a high level of uncertainty in the agricultural production network. In relation to crop irrigation, it enables and improves food production, but encompasses several factors that require techniques and the proper management of irrigation water. Therefore, this work proposes the development of an IoT-based architecture to optimize the use of irrigation, introducing an ML-trained recommendation system that analyzes each plant individually, in order to catalog the necessary conditions for its healthy growth. The expected results are the creation of an architecture for automated irrigation systems that consists of creating a dataset from sensors connected and configured in an IoT module and a virtual weather station that will generate models for machine learning training.