Faculty Contact: TBN
Abstract: Recently, there has been an explosion of interest in precision agriculture as the need for food stability moves to the forefront of global issues. Various techniques and systems have been developed by both the academic and business worlds, including in-situ sensors, satellite imaging, and UAV monitoring. Thanks to the affordability of things like sensor hardware and cloud storage/computation, driven largely by the growing popularity of IoT (Internet of Things), these systems for the first time seem practical for large-scale deployment. However, several issues remain in the way of this reality. When deploying in-situ sensors -- with or without complementing techniques -- each system must be customized to achieve optimum results. Network topology is of utmost importance when considering sensor placement, both for data collection purposes as well as for connectivity. The goal of this module is to develop a framework which will streamline, standardize, and optimize the implementation of in-situ sensor systems for agricultural landscapes, to the point where individual (and/or small-scale) farmers can independently tailor this themselves to suit their needs. Hyperspectral satellite imagery will be used in combination with graph theoretic approaches to give practical sensor-placement suggestions given various geographic, capital, and connectivity constraints. We will test a physical implementation of this framework with LoRa™ IoT sensors.
- Spring 2017: Haleigh Wright and Lilla Bartko