[Paper Review] Real-time Geoinformation Systems to Improve the Quality, Scalability, and Cost of Internet of Things for Agri-environment Research

Bryan C. Runck, Bobby Schulz+42 more|arXiv (Cornell University)|Mar 28, 2024

Smart Agriculture and AIAgricultural and Biological Sciences3 citations

TL;DR

This paper presents an open-source, real-time geoinformation system for agricultural and environmental research that enhances data quality, scalability, and cost efficiency in IoT deployments. By integrating open-source hardware, standardized software, and a sustainable public-private business model, the system enables large-scale, interoperable sensor networks across four continents, supporting AI-driven research through reliable, high-resolution spatiotemporal data pipelines.

ABSTRACT

With the increasing emphasis on machine learning and artificial intelligence to drive knowledge discovery in the agricultural sciences, spatial internet of things (IoT) technologies have become increasingly important for collecting real-time, high resolution data for these models. However, managing large fleets of devices while maintaining high data quality remains an ongoing challenge as scientists iterate from prototype to mature end-to-end applications. Here, we provide a set of case studies using the framework of technology readiness levels for an open source spatial IoT system. The spatial IoT systems underwent 3 major and 14 minor system versions, had over 2,727 devices manufactured both in academic and commercial contexts, and are either in active or planned deployment across four continents. Our results show the evolution of a generalizable, open source spatial IoT system designed for agricultural scientists, and provide a model for academic researchers to overcome the challenges that exist in going from one-off prototypes to thousands of internet-connected devices.

Motivation & Objective

• Address the scalability and data quality challenges in deploying large-scale, real-time IoT systems for agricultural and environmental research.
• Overcome the 'IoT scaling gap' that hinders the transition from prototype sensors to production-grade, field-deployed systems.
• Develop a sustainable, open-source ecosystem for geospatial IoT that supports academic research and public-private partnerships.
• Enable interoperability and long-term data integrity through standardized software, databases, and data quality assurance workflows.
• Create a cost-effective, institutionally supported model for deploying thousands of internet-connected sensors across diverse research domains.

Proposed method

• Design and deploy an open-source spatial IoT stack comprising custom hardware loggers, firmware, and cloud-based data pipelines for real-time data ingestion.
• Implement real-time data quality control mechanisms to detect and correct errors during data collection, ensuring analysis-ready datasets.
• Adopt a modular, version-controlled software architecture with 17 total system versions (3 major, 14 minor), supporting iterative development and deployment.
• Integrate with institutional infrastructure including the University of Minnesota’s Supercomputing Institute and DRUM for long-term data archiving.
• Establish a dual-tier business model: cost-based pricing for internal users and subsidized, revenue-generating pricing for external partners to support R&D sustainability.
• Use technology readiness level (TRL) framework to guide system evolution from prototype to mature, field-deployed applications across multiple research domains.

Experimental results

Research questions

• RQ1 How can open-source geospatial IoT systems be designed to ensure high data quality and real-time processing at scale in agricultural research?
• RQ2 What institutional and business models are effective for sustaining long-term deployment of open-source IoT systems in academic research environments?
• RQ3 To what extent can standardized software and hardware components improve interoperability and reduce development costs in multi-domain agri-environmental sensing?
• RQ4 How does integrating real-time data quality assurance into IoT pipelines enhance the reliability and utility of AI/ML training datasets?
• RQ5 What role do public-private partnerships play in scaling open-source IoT systems for digital agriculture beyond academic prototypes?

Key findings

• The GEMS Sensing system successfully deployed over 2,727 IoT devices across four continents, demonstrating scalability in real-world agricultural and environmental research settings.
• The system achieved high data quality through real-time validation and error detection, reducing manual data cleaning and improving dataset reliability for AI/ML applications.
• The open-source hardware and firmware design enabled cost-effective manufacturing and rapid iteration, with devices produced in both academic and commercial settings.
• The dual business model—cost-based for internal users and subsidized for external partners—supported sustainable R&D and enabled public-private collaboration for long-term system viability.
• Integration with institutional infrastructure (e.g., Minnesota Supercomputing Institute, DRUM, GEMS Platform) ensured data persistence, discoverability, and reproducibility across research lifecycles.
• The system’s evolution through 17 versions (3 major) demonstrated the feasibility of a modular, version-controlled approach to building robust, production-grade IoT systems for science.