Acquisition and Analysis of Soil Parameters for Vine Monitoring using the IoT Sensor Network

In recent years, the concept of the Internet of Things (IoT) has spread in most fields due to the multiple benefits it offers, so it recently started to be used in viticulture, together with artificial intelligence (AI). These two innovative technologies combined with classical methods can lead to a much more efficient and rapid prediction of diseases that may occur in grapevine culture. The present study aims to detect diseases in the period when evolution can still be prevented, by minimizing the favorable environment in which the pathogen can develop. This prevention can be achieved using algorithms for predicting and comparing the data from IoT sensors with the data classically gathered by farmers. The IoT sensors system developed to acquire information about environmental parameters is tested in the Murfatlar vineyard, Romania. Two types of untreated Cabernet and Sauvignon Blanc varieties are used in the experimental study with the aim of reducing the chemical treatment concentration. All the sensors’ data are stored in a cloud. This paper presents an analysis of sensor data consisting of soil parameters such as temperature, humidity, conductivity, nitrogen (N), phosphorus (P), and potassium (K). Big data algorithms using machine learning (ML) for clustering and prediction are tested to identify the vine diseases. The results are detailed at the end of the paper.