Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Comparing Machine Learning Techniques for Throughput Estimation using Partial Data

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2021 ASABE Annual International Virtual Meeting  2100791.(doi:10.13031/aim.202100791)
Authors:   Steffen Schwich, Jan Schattenberg, Dr.-Ing., Ludger Frerichs, Prof. Dr.
Keywords:   Deep-Learning, LSTM, Machine Learning, Yield Estimation, Throughput

Abstract. In this paper a deep-learning-based method for a continuous throughput calculation using mechanical metrics of the harvester is developed. The algorithm is able to predict the yield with a mean error of approx. 2% on a validation and test dataset. Since the yield is calculated by the integration of the throughput, the results enable the creation of yield maps for the future, which than can enhance the smart farming in term of the sugar beet production.

(Download PDF)    (Export to EndNotes)