DeepPaSTL: Spatio-Temporal Deep Learning Methods for Predicting Long-Term Pasture Terrains Using Synthetic Datasets

Effective management of dairy farms requires an accurate prediction of pasture biomass. Generally, estimation of pasture biomass requires site-specific data, or often perfect world assumptions to model prediction systems when field measurements or other sensory inputs are unavailable. However, for s...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Murtaza Rangwala, Jun Liu, Kulbir Singh Ahluwalia, Shayan Ghajar, Harnaik Singh Dhami, Benjamin F. Tracy, Pratap Tokekar, Ryan K. Williams
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	agriculture convolution neural network prediction remote sensing recurrent sequence biomass Agriculture S
Acceso en línea:	https://doaj.org/article/385b740902a5406b8653f4fb755e6b41
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:385b740902a5406b8653f4fb755e6b41
record_format	dspace
spelling	oai:doaj.org-article:385b740902a5406b8653f4fb755e6b412021-11-25T16:08:26ZDeepPaSTL: Spatio-Temporal Deep Learning Methods for Predicting Long-Term Pasture Terrains Using Synthetic Datasets10.3390/agronomy111122452073-4395https://doaj.org/article/385b740902a5406b8653f4fb755e6b412021-11-01T00:00:00Zhttps://www.mdpi.com/2073-4395/11/11/2245https://doaj.org/toc/2073-4395Effective management of dairy farms requires an accurate prediction of pasture biomass. Generally, estimation of pasture biomass requires site-specific data, or often perfect world assumptions to model prediction systems when field measurements or other sensory inputs are unavailable. However, for small enterprises, regular measurements of site-specific data are often inconceivable. In this study, we approach the estimation of pasture biomass by predicting sward heights across the field. A convolution based sequential architecture is proposed for pasture height predictions using deep learning. We develop a process to create synthetic datasets that simulate the evolution of pasture growth over a period of 30 years. The deep learning based pasture prediction model (DeepPaSTL) is trained on this dataset while learning the spatiotemporal characteristics of pasture growth. The architecture purely learns from the trends in pasture growth through available spatial measurements and is agnostic to any site-specific data, or climatic conditions, such as temperature, precipitation, or soil condition. Our model performs within a 12% error margin even during the periods with the largest pasture growth dynamics. The study demonstrates the potential scalability of the architecture to predict any pasture size through a quantization approach during prediction. Results suggest that the DeepPaSTL model represents a useful tool for predicting pasture growth both for short and long horizon predictions, even with missing or irregular historical measurements.Murtaza RangwalaJun LiuKulbir Singh AhluwaliaShayan GhajarHarnaik Singh DhamiBenjamin F. TracyPratap TokekarRyan K. WilliamsMDPI AGarticleagricultureconvolution neural networkpredictionremote sensingrecurrent sequencebiomassAgricultureSENAgronomy, Vol 11, Iss 2245, p 2245 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	agriculture convolution neural network prediction remote sensing recurrent sequence biomass Agriculture S
spellingShingle	agriculture convolution neural network prediction remote sensing recurrent sequence biomass Agriculture S Murtaza Rangwala Jun Liu Kulbir Singh Ahluwalia Shayan Ghajar Harnaik Singh Dhami Benjamin F. Tracy Pratap Tokekar Ryan K. Williams DeepPaSTL: Spatio-Temporal Deep Learning Methods for Predicting Long-Term Pasture Terrains Using Synthetic Datasets
description	Effective management of dairy farms requires an accurate prediction of pasture biomass. Generally, estimation of pasture biomass requires site-specific data, or often perfect world assumptions to model prediction systems when field measurements or other sensory inputs are unavailable. However, for small enterprises, regular measurements of site-specific data are often inconceivable. In this study, we approach the estimation of pasture biomass by predicting sward heights across the field. A convolution based sequential architecture is proposed for pasture height predictions using deep learning. We develop a process to create synthetic datasets that simulate the evolution of pasture growth over a period of 30 years. The deep learning based pasture prediction model (DeepPaSTL) is trained on this dataset while learning the spatiotemporal characteristics of pasture growth. The architecture purely learns from the trends in pasture growth through available spatial measurements and is agnostic to any site-specific data, or climatic conditions, such as temperature, precipitation, or soil condition. Our model performs within a 12% error margin even during the periods with the largest pasture growth dynamics. The study demonstrates the potential scalability of the architecture to predict any pasture size through a quantization approach during prediction. Results suggest that the DeepPaSTL model represents a useful tool for predicting pasture growth both for short and long horizon predictions, even with missing or irregular historical measurements.
format	article
author	Murtaza Rangwala Jun Liu Kulbir Singh Ahluwalia Shayan Ghajar Harnaik Singh Dhami Benjamin F. Tracy Pratap Tokekar Ryan K. Williams
author_facet	Murtaza Rangwala Jun Liu Kulbir Singh Ahluwalia Shayan Ghajar Harnaik Singh Dhami Benjamin F. Tracy Pratap Tokekar Ryan K. Williams
author_sort	Murtaza Rangwala
title	DeepPaSTL: Spatio-Temporal Deep Learning Methods for Predicting Long-Term Pasture Terrains Using Synthetic Datasets
title_short	DeepPaSTL: Spatio-Temporal Deep Learning Methods for Predicting Long-Term Pasture Terrains Using Synthetic Datasets
title_full	DeepPaSTL: Spatio-Temporal Deep Learning Methods for Predicting Long-Term Pasture Terrains Using Synthetic Datasets
title_fullStr	DeepPaSTL: Spatio-Temporal Deep Learning Methods for Predicting Long-Term Pasture Terrains Using Synthetic Datasets
title_full_unstemmed	DeepPaSTL: Spatio-Temporal Deep Learning Methods for Predicting Long-Term Pasture Terrains Using Synthetic Datasets
title_sort	deeppastl: spatio-temporal deep learning methods for predicting long-term pasture terrains using synthetic datasets
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/385b740902a5406b8653f4fb755e6b41
work_keys_str_mv	AT murtazarangwala deeppastlspatiotemporaldeeplearningmethodsforpredictinglongtermpastureterrainsusingsyntheticdatasets AT junliu deeppastlspatiotemporaldeeplearningmethodsforpredictinglongtermpastureterrainsusingsyntheticdatasets AT kulbirsinghahluwalia deeppastlspatiotemporaldeeplearningmethodsforpredictinglongtermpastureterrainsusingsyntheticdatasets AT shayanghajar deeppastlspatiotemporaldeeplearningmethodsforpredictinglongtermpastureterrainsusingsyntheticdatasets AT harnaiksinghdhami deeppastlspatiotemporaldeeplearningmethodsforpredictinglongtermpastureterrainsusingsyntheticdatasets AT benjaminftracy deeppastlspatiotemporaldeeplearningmethodsforpredictinglongtermpastureterrainsusingsyntheticdatasets AT prataptokekar deeppastlspatiotemporaldeeplearningmethodsforpredictinglongtermpastureterrainsusingsyntheticdatasets AT ryankwilliams deeppastlspatiotemporaldeeplearningmethodsforpredictinglongtermpastureterrainsusingsyntheticdatasets
_version_	1718413309719347200

DeepPaSTL: Spatio-Temporal Deep Learning Methods for Predicting Long-Term Pasture Terrains Using Synthetic Datasets

Ejemplares similares