Prediction of non-linear time-variant dynamic crop model using Bayesian methods

Abstract

This work addresses the problem of predicting a non-linear time-variant leaf area index and soil moisture model (LSM) using state estimation. These techniques include the extended Kalman filter (EKF), particle filter (PF) and the more recently developed technique, variational filter (VF). In the comparative study, the state variables (the leaf-area index LAI, the volumetric water content of the layer 1, HUR1 and the volumetric water content of the layer 2, HUR2) are estimated from noisy measurements of these variables, and the various estimation techniques are compared by computing the estimation root mean square error with respect to the noise-free data. The results show that VF provides a significant improvement over EKF and PF.