Automated Irrigation System Based on LoRa and ML for Marginal Farmers

Abstract

India is one of the foremost agricultural producers in the world; on the other hand, the consumption of water for agricultural purposes in India has been among the highest in the world. Indiscriminate use of inadequate irrigation techniques has led to a critical water deficit in the country. Now with the development of Internet of Things (IoT) Precision Farming and Precision Irrigation are becoming very popular. This paper proposes a cost-effective Automated Irrigation System based on LoRa and Machine Learning, which can be of great help to marginal farmers, for whom agriculture is hardly a profitable venture, mainly due to water scarcity. In this automated system, LoRa technology is used in Sensor and Irrigation node, in which sensors collect data on soil moisture and temperature and send it to the server through a LoRa gateway. Then the data is fed into a Machine Learning algorithm, which leads to correct prediction of the soil status. Hence, the field needs to be irrigated only if and when it is needed. The system can be remotely monitored using a web application that can be accessed by a mobile phone.

A Survey on LoRa for Smart Agriculture: Current Trends and Future Perspectives

“…makes possible to develop and train specific irrigation models to calculate the exact amount of water to be distributed. For example, the works [54], [55], [56] provide machine learning-based smart irrigation systems, all employing LoRa technology. In particular, in [55] a Random Forest Classifier predicts the soil moisture and thus irrigation is planned accordingly.…”

“…The energy-neutral irrigation node described in [51]. [43] x x x x x x Kodali et al [44] x x x x x x Nisa et al [45] x x x x x x Ali et al [46] x x x x x x x x x Pump current Jiang et al [47] x x x x x Zhang et al [48] x x x x x Pressure Emharraf et al [49] x x x Battery level Yuan et al [50] x x x x Zhao et al [51] x x x x x x x x x Usmonov et al [52] x x x x x x x x x x Vu et al [53] x x x x x Chang et al [54] x x x Henna et al [55] x x x x Pressure Loganathan et al [56] x x x x x x Froiz-Mìguez et al [57] x…”