An IoT-based Smart Agriculture System with Locust Prevention and Data Prediction

Abstract

Locust and grasshopper infestation have a long history of affecting crops and human lives. From ancient Egypt to the Bronze age, everywhere, we have seen the manifestation of locust outbreaks and how humans have fought against it for their survival generations after generations. The latest locust eruption began in June 2019 and has continued through 2020. It has been the worst one in the last 70 years in Middle Africa, Middle East, South Asia, and South America. Countries are taking precautions to be safe from this outbreak because, after this corona pandemic, no nation is willing to face another economic pandemic. In advances of facing the consequences of the locust swarms, we need to find an effective and smart solution. In this paper, we have come up with the idea of monitoring important agricultural factors such as soil moisture, temperature, and humidity using sensors to provide real-time information to the farmers about imminent locust infestation to their mobile. Also, to ease their work, our proposed system will provide water and pesticides automatically to the fields by using Raspberry Pi and Node MCU. Our proposed system will generate ultraviolet light and loud noise to kill the insects in case of a locust outbreak. As locust's habitats are closely related to different agricultural factors, linear regression, logistic regression, and support vector regression, machine learning algorithms have been implemented to predict the temperature and humidity so that the farmers can anticipate these factors well ahead of time and plan accordingly. Overall a next-generation solution to fight the locusts has been implemented in this paper.

Topics & Concepts