Green IoT and Machine Learning for Agricultural Applications

Abstract

The Internet of Things (IoT) is a network of various sensors, software, and other technologies embedded into a system retaining flexibility and feasibility in human work. This involves the framework of various sensors and software in the form of an embedded system to perform either a single or multiple tasks simultaneously. IoT is a continuously building network of smart sensors and devices that are connected to the internet. However, the connected IoT devices require a huge amount of energy for high and efficient performance to provide a sophisticated environment to the users. This had become a huge concern and gained a huge focus on the upcoming research providing a pathway for the G-IoT, i.e., Green IoT. G-IoT represents the framework of connecting smart sensors and devices and creating automation by enabling energy conservation methods. Cloud Computing and Machine Learning techniques like Artificial Neural Networks, C-Means, K-Means, and Bayesian Model are used for creating IoT systems that are cost-effective and has low-power consumption. Energy consumption depends upon the system architecture and its requirements. The approaches like Green Computing and Green Wireless Sensor Networks make difference in the energy consumption criteria. G-IoT has many applications in agriculture, namely, smart agribots, machine navigation, harvesting robots, smart farming kits, material handling, agricultural drones, remote sensing for crop and weather conditions, monitoring and sensing soil quality, computer imaging techniques for quality control, sorting, and grading, and irrigation monitoring. Moreover, Machine Learning had become an emerging paradigm and the tools or models are quite widely used in various applications. In agriculture, Machine Learning tools are used for species management, field conditions management, crop management, and livestock management.