The effectiveness of IoT and machine learning in Precision Agriculture

Applying the IoT paradigm in agriculture generates an unprecedented amount of data about fields and crops worldwide. With data, machine learning is natural to forecast or estimate crop parameters, which is remarkable for the estimation of evapotranspiration and soil moisture to propose irrigation plans and control, among other applications. This work investigates the literature to evaluate what kind of IoT devices and machine learning algorithms are being applied in precision agriculture and discusses the effectiveness of such equipment and algorithms on different spatial scales. The results indicate IoT devices are commonly applied in small crop areas or greenhouses and for short periods of time, usually less than one growing season. Machine learning models are also applied in similar scenarios, and the effectiveness of such techniques is discussed according to growers' needs and ambitions.