Performance Triggered Adaptive Model Reduction for Soil Moisture Estimation in Precision Irrigation

ABSTRACT Accurate soil moisture information is essential for precise irrigation to enhance water use efficiency. Estimating soil moisture based on limited soil moisture sensors is especially critical for obtaining comprehensive soil moisture information when dealing with large‐scale agricultural fields. The major challenge in soil moisture estimation lies in the high dimensionality of the spatially discretized agro‐hydrological models. In this work, we propose a performance‐triggered adaptive model reduction approach to address this challenge. The proposed approach employs a prediction performance‐based triggering scheme to activate model updates adaptively in a way such that the prediction error between the reduced model and the original model over a prediction horizon is maintained below a predetermined threshold. For each update to the model, a trajectory‐based unsupervised machine learning technique is used. An adaptive extended Kalman filter (EKF) is designed based on the reduced model for soil moisture estimation. The applicability and performance of the proposed approach are extensively evaluated through the application to a simulated large‐scale agricultural field.