RuralAI in Tomato Farming: Integrated Sensor System, Distributed Computing, and Hierarchical Federated Learning for Crop Health Monitoring

Journal Article · 03 April 2024 · IEEE Sensors Letters

DOI: https://doi.org/10.1109/lsens.2024.3384935 · OSTI ID:2473153

Devaraj, Harish  ^[1]; Sohail, Shaleeza  ^[2]; Ooi, Melanie  ^[1]; Li, Boyang  ^[2]; Hudson, Nathaniel  ^[3]; Baughman, Matt  ^[3]; Chard, Kyle  ^[3]; Chard, Ryan  ^[4]; Casella, Enrico  ^[5]; Foster, Ian  ^[3]; Rana, Omer  ^[6]

---

1. Univ. of Waikato, Hamilton (New Zealand)
2. University of Newcastle, Callaghan, NSW (Australia)
3. Univ. of Chicago, IL (United States)
4. Argonne National Laboratory (ANL), Argonne, IL (United States)
5. Univ. of Wisconsin, Madison, WI (United States)
6. Cardiff Univ., Wales (United Kingdom)

Precision horticulture is evolving due to scalable sensor deployment and machine learning (ML) integration. These advancements boost the operational efficiency of individual farms, balancing the benefits of analytics with autonomy requirements. However, given concerns that affect wide geographic regions (e.g., climate change), there is a need to apply models that span farms. Federated learning (FL) has emerged as a potential solution. FL enables decentralized ML across different farms without sharing private data. Traditional FL assumes simple two-tier network topologies and, thus, falls short of operating on more complex networks found in real-world agricultural scenarios. Networks vary across crops and farms and encompass various sensor data modes, extending across jurisdictions. New hierarchical FL (HFL) approaches are needed for more efficient and context-sensitive model sharing, accommodating regulations across multiple jurisdictions. Here, we present the RuralAI architecture deployment for tomato crop monitoring, featuring sensor field units for soil, crop, and weather data collection. HFL with personalization is used to offer localized and adaptive insights. Model management, aggregation, and transfers are facilitated via a flexible approach, enabling seamless communication between local devices, edge nodes, and the cloud.

• Research Organization: Argonne National Laboratory (ANL), Argonne, IL (United States)
• Sponsoring Organization: USDOE Office of Science (SC), Basic Energy Sciences (BES)
• Grant/Contract Number: AC02-06CH11357
• OSTI ID: 2473153
• Journal Information: IEEE Sensors Letters, Journal Name: IEEE Sensors Letters Journal Issue: 5 Vol. 8; ISSN 2475-1472
• Publisher: IEEECopyright Statement
• Country of Publication: United States
• Language: English