Many proposed AgTech solutions do not take into account the environment of agriculture production. This project aims to develop, characterize, and test in real-use conditions, novel inexpensive wireless sensors for accurately measuring agriculturally-relevant soil signals such as nitrate, ammonium, phosphate, potassium, and moisture. In addition, AI-enabled algorithms will be developed to predict the spatiotemporal distribution of these signals to support 4R Nutrient Stewardship practices.

This project will be a joint effort of UC Berkeley and UC Davis researchers, with the novel soil sensors developed at UCB using printing technologies and conventional electronics; and the sensors’ performance evaluated at UCD, under controlled laboratory conditions and in instrumented commercial almond orchard and processing tomato fields. To do this, a UAV will be outfitted with a wireless readout system and programmed at UCB to navigate the field for sampling data from the sensors and uploaded to the cloud from its base station. In addition, supervised learning AI algorithms for spatiotemporal prediction of soil analytes will be developed at UCD. With rigorous testing and evaluation methods, the team will deliver affordable instrumentation for real-time soil monitoring and AI-enabled algorithms that will enhance production, profitability, and environmental stewardship.