Physics-Informed Neural Network Methods for Predicting Plant Height Development

Abstract

ABSTRACT Plant growth is a dynamic process affected by genes and growing environment, with all kinds of interactions between them. These complex relationships make the prediction of plant growth challenging. We propose a hybrid modelling framework that combines a logistic ordinary differential equation model with a Long Short-Term Memory (LSTM) neural network, resulting in a Physics Informed Neural Network (PINN). While PINNs have been widely applied to physical dynamical systems, their use in modelling the dynamics of plant growth systems is still largely unexplored. We illustrate the construction of a PINN on plant height data in wheat and compare its performance with alternative models for longitudinal plant data. All temporal prediction models only require time and temperature as input. Among a set of competing models, our PINN had the lowest average root mean squared error (RMSE) of prediction and the smallest standard deviation across multiple random initialisations. Therefore, we conclude that incorporating biological growth constraints into data-driven growth models can enhance prediction accuracy of longitudinal plant traits. Highlights Integrating plant growth equations into a temporal neural network improves plant height growth prediction over ordinary differential equations and machine learning models, especially when training data are limited.

Plant phenotyping relevance

植物高の時系列形質を予測するPINNを開発し、代替モデルと精度比較しているため、植物表現型の計算手法が中心である。

Code and data availability

The supplied blocks describe a PINN/LSTM plant-height model trained on ETH Zürich FIP wheat data (Roth et al., 2025), but contain no authors' public code, data deposit, or availability statement with a URL. The FIP dataset is cited prior work, not a paper-specific asset. No qualifying assets are present.

No evidence-backed public reproduction asset is currently recorded.