AgML: Open-Source Machine Learning Infrastructure Pilot Project with Optimized Sensor-Driven Resilient Precision Agriculture
Year 2 Research Project

AgML: Open-Source Machine Learning Infrastructure Pilot Project with Optimized Sensor-Driven Resilient Precision Agriculture

Description

SIGNIFICANCE

The three teams on this project are collaborating in the areas of open-source infrastructure development, optimized sensor-driven precision agriculture, and novel inexpensive wireless sensors for accurately measuring soil nitrate. The proposed open-source infrastructure development project has the potential to accelerate the scaling of agricultural AI technologies across academia and industry. Currently, no such framework exists for agricultural AI. The proposed open-source Python library, called AgML, will provide data and code resources to academic and industry ML developers, ultimately aiming to build a broader open-source community through shared infrastructure. The proposed optimized sensor-driven resilient precision agriculture component of the project could ultimately result in decreasing economic damage from airborne delivery of agricultural products. Additionally, the addition of inexpensive soil nitrate sensors could reduce cost and labor involved in measuring soil nitrate levels.

GOALS

AgML’s objectives are to centralize and standardize datasets, develop benchmarks and pre-trained models, create ag-specific ML workflows, and generate synthetic ML agricultural data. The optimized sensor work will provide useful tools to enable rapid path planning for autonomous vehicle operators in real-time and to train operators in large surface area food systems. The soil nitrate sensors will be developed and characterized in real world conditions.

Photos

AgML: Open-Source Machine Learning Infrastructure Pilot Project with Optimized Sensor-Driven Resilient Precision Agriculture
AgML: Open-Source Machine Learning Infrastructure Pilot Project with Optimized Sensor-Driven Resilient Precision Agriculture
AgML: Open-Source Machine Learning Infrastructure Pilot Project with Optimized Sensor-Driven Resilient Precision Agriculture
AgML: Open-Source Machine Learning Infrastructure Pilot Project with Optimized Sensor-Driven Resilient Precision Agriculture
AgML: Open-Source Machine Learning Infrastructure Pilot Project with Optimized Sensor-Driven Resilient Precision Agriculture
AgML: Open-Source Machine Learning Infrastructure Pilot Project with Optimized Sensor-Driven Resilient Precision Agriculture

Team

Portrait of Mason Earles

Mason Earles

Principal Investigator

Portrait of Tarek Zohdi

Tarek Zohdi

Principal Investigator

Portrait of Isaya Kisekka

Isaya Kisekka

Principal Investigator

Portrait of Zhaodan Kong

Zhaodan Kong

Collaborator

Portrait of Brian Bailey

Brian Bailey

Collaborator

Portrait of Stavros Vougioukas

Stavros Vougioukas

Collaborator

Portrait of Alireza Pourreza

Alireza Pourreza

Collaborator

Portrait of Yufang Jin

Yufang Jin

Collaborator

Portrait of Ana Arias

Ana Arias

Collaborator

Portrait of Ethan Ligon

Ethan Ligon

Collaborator

Portrait of Xin Liu

Xin Liu

Collaborator

Portrait of Mark Mueller

Mark Mueller

Collaborator

Portrait of Ilias Tagkopoulos

Ilias Tagkopoulos

Collaborator

Publications

Journal Article · Plant Phenomics · 2023

Standardizing and Centralizing Datasets for Efficient Training of Agricultural Deep Learning Models

Joshi, Amogh, Dario Guevara, and Mason Earles