Edge computing in Smart Agriculture scenario based on TinyML for irrigation control

Various regions around the world are encountering substantial difficulties as a result of arid conditions, vulnerability to climate change, and water shortages. These factors present significant obstacles to maintaining agricultural practices. Water stress resulting from limited water availability adversely affects crop growth and productivity. To tackle this pressing issue, the application of Tiny Machine Learning (TinyML) models for optimizing crop irrigation is proposed. TinyML involves implementing machine learning algorithms on resource-constrained devices, such as microcontrollers. By collecting and processing meteorological and crop data, a TinyML model has been trained to optimize irrigation practices and detect crop anomalies in real-time. This approach has the potential to revolutionize agricultural practices by enabling precise and efficient water management, even in remote environments. This work tackles the first steps to implementing TinyML algorithms for irrigation. We have computed the relationship between humidity and temperature changes with NDVI (Normalized Difference Vegetation Index) anomalies in crops to make predictions and make intelligent decisions toward water usage.