Soil AI Lab
Soil AI Lab

Welcome to Our Lab.

We develop trustworthy AI to democratize soil, crop, and environmental intelligence from digital data streams, delivering clear and usable insights for agriculture and beyond.

We are part of the University of Florida Institute of Food and Agricultural Sciences (UF/IFAS).

Soil Science & Artificial Intelligence

Our Team

OUR MISSION

The Soil AI Lab was created to advance a simple idea: the ground beneath our feet is one of the most important, and most overlooked, systems on Earth. Soil regulates food production, water quality, carbon storage, and ecosystem resilience, yet it remains difficult to measure, monitor, and manage at the scales required for smart decision making in agriculture.

Our mission close that gap by combining soil science, Earth observation, advanced sensing, and trustworthy artificial intelligence into practical tools that make soil and crop information more accessible, timely, and actionable for everyone.

Field Spectroscopy for Rapid Soil Diagnostics

We use portable spectroscopy in the lab and field to quickly measure soil conditions without destructive sampling.

Autonomous Robotics

We develop and deploy field robots to collect repeatable, geo-tagged soil related information.

Hyperspectral Remote Sensing & Data Fusion

We combine satellite, UAV, and ground data to monitor soils and crops consistently across scales.

GeoAI Decision Tools

We develop smart AI tools, leveraging large language models, to simplify access to geospatial insights and support faster, more informed decisions.

Explainable AI

We develop interpretable models with uncertainty estimates and decentralized systems so users can trust and apply the results.

Open Science & Community Engagement

We promote open science by hosting working groups and live demonstrations, and by sharing open data and codewhenever possible.

HISTORY
Founded from scratch in March 2023 in a new space, the lab has grown steadily into an international team with diverse backgrounds and global collaborations. We combine strong access to high-performance computing with state-of-the-art spectral sensors and cameras, enzymatic readers, and low-cost robotic platforms (including legged robots and UAVs), alongside trusted traditional field and lab equipment. This integrated toolkit supports scalable field campaigns and strengthens data quality, validation, and real-world deployment—from fields next to us in Florida to the tropical Amazon in Brazil, and arid landscapes across the Mediterranean and Australia.

Our research has expanded into trustworthy AI and multi-source data fusion to produce robust crop-soil indicators and decision-ready products. Today, we build scalable tools such as interactive viewers, and chat-map AI interfaces that help users move from data to decisions faster, with transparency and quantified uncertainty.