Design of an all-in-one sensing node for climate smart agriculture scenarios

The integration of advanced technologies into agricultural practices, often referred to as precision agriculture, represents a scientific and engineering approach to optimizing farming operations, improving productivity, and promoting sustainability. This thesis is part of the IRIDE project, an interregional initiative primarily aimed at conserving water in irrigation systems and providing automated data-driven support to farmers. The core of the thesis work involves the hardware design of a device dedicated to collecting sensor data related to water content of the soil, environmental conditions and health status of the cultivated plants. The deployment of embedded electronic systems directly in agricultural fields and orchards enables the collection of accurate measurements and allows the implementation of sophisticated algorithms for the monitoring and prediction of plant health status. Assuming an ideal prediction of water-deficit stress conditions of the plants, irrigation could be applied at the most appropriate time and with the optimal amount of water. As a result, crop yields could be maintained or even improved without negatively impacting community water reserves or groundwater aquifers. The IRIDE project requires a low-power, long-range, and cost-effective radio communication network for remote sensing in agricultural fields. Accordingly, this thesis leverages a PCB developed by the supervising research group, referred to as the LoRaTO board, to implement the LoRaWAN protocol stack over a commercial IoT infrastructure. The resulting sensing node enables data collection of soil parameters (volumetric water content, soil temperature and matric potential), environmental parameters (gas concentration, air temperature, pressure and humidity), water flow within irrigation pipes, and plant-stem hydration status. The latter is achieved by measuring the frequency of a relaxation oscillator whose feedback loop incorporates a segment of the cultivated plant stem and other components for automatic calibration of the frequency range. The system is battery-powered and includes fault-prevention mechanisms against overcurrent, overtemperature, reverse polarity and unexpected discharge. Additionally, its LoRaWAN connection supports remote firmware updates using the LoRaWAN FUOTA process. Special attention was given throughout the design process—from component selection to board layout—to address energy efficiency and minimize the cost and size of the sensing node. The system also supports scalability across different agricultural environments, thanks to its variable number of connected sensors and its compatibility with various plant species. In addition, the design enables straightforward maintenance by non-specialized personnel and is broadly compatible with field, greenhouse, orchard, and laboratory applications, allowing medium-sized companies to benefit from precision farming without substantial capital investment or highly trained staff. The final phase of the thesis work, after the fabrication and assembly of the system, consisted of preliminary laboratory tests performed to verify and validate its functionality.