Weather-Driven Agricultural Decision-Making Using Digital Twins Under Imperfect Conditions
This work addresses data quality issues for agricultural automation, though it appears incremental as it applies existing digital twin and anomaly detection methods to this specific domain.
The researchers tackled the problem of imperfect weather data in agricultural decision-making by developing Cerealia, a modular digital twin framework that detects data inconsistencies using neural network models, achieving operational testing in a commercial orchard and with public datasets.
By offering a dynamic, real-time virtual representation of physical systems, digital twin technology can enhance data-driven decision-making in digital agriculture. Our research shows how digital twins are useful for detecting inconsistencies in agricultural weather data measurements, which are key attributes for various agricultural decision-making and automation tasks. We develop a modular framework named Cerealia that allows end-users to check for data inconsistencies when perfect weather feeds are unavailable. Cerealia uses neural network models to check anomalies and aids end-users in informed decision-making. We develop a prototype of Cerealia using the NVIDIA Jetson Orin platform and test it with an operational weather network established in a commercial orchard as well as publicly available weather datasets.