IoT-Based Soil Monitoring System for Soil Strength and Classification Using Machine Learning

This paper proposes the design of an innovative remote soil monitoring system to measure and classify the soil strength and properties using soil resistivity and other soil parameters. The proposed system consists of a microcontroller with multiple sensors, including humidity, pH and rain sensor along with the Wenner method of soil resistivity measurement. Currently in our proposed system we tested with Arduino and ESP 32 microcontrollers. The proposed automated soil resistivity calculation opens up more advantages than the conventional ones. For the transmission of data, this system proposing a hybrid approach combining different protocols. On the receiver end the system includes receiver and analysing unit, the analysis unit uses a machine learning algorithm to interpret the collected data to classify soil types and strength. This paper also proposing a suitable machine learning algorithm for an IoT-based soil monitoring system depending on the type of data and the task at hand. Application of this system includes environmental management and precision agriculture. It also provides real time monitoring and decision making thereby supporting improvement of geotechnical Practices and optimization of agriculture processes.