Assessment of sensor driven automatic smart soil and paddy seed metering mechanisms using artificial intelligence for paddy nurseries
Date
2025
Authors
Choudhary, V.
Machavaram, R.
Patidar, P.
Singh, G.
Singh, N.
Kumawat, L.
Editors
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Journal article
Citation
Smart Agricultural Technology, 2025; 10:1-16
Statement of Responsibility
Conference Name
Abstract
A sensor-based automatic soil and paddy seed metering mechanism is essential to ensure uniform distribution of soil and seeds in trays with precise quantities, optimizing seedling growth and efficiency. This is important for fostering optimal biometric seedling growth and ensuring sustainable paddy production. Thus, in present study, a sensor based metering mechanism for soil filling and paddy seeding was developed and evaluated.
The operating and metering parameters included the speed of the chain conveyor (0.123 to 0.27 m/s), the speed of the base soil feed unit (539 to 718 rpm), the speed of the paddy seed feed unit (43 to 53 rpm), and the speed of the topsoil feed unit (108 to 180 rpm). Optimizing these parameters in tray nurseries to achieve a base soil depth of 17.5 mm, a paddy seed quantity of 150 g per tray, and a topsoil depth of 4 mm is necessary to ensure uniform distribution. The optimized values of the speed of the chain conveyor, speed of the base soil feed unit, speed of the paddy seed feed unit, and speed of the topsoil feed unit were obtained to be 0.21 m/s, 651.45 rpm, 52.99 rpm, and 150.40 rpm, respectively.
At the optimized condition, the values of depth of base soil, the quantity of paddy seed, depth of topsoil, coefficient of base soil distribution uniformity, coefficient of paddy seed distribution uniformity, and coefficient of topsoil distribution uniformity were predicted as 15.97 mm, 143.73 g/tray, 3.41 mm, 97.53 %, 96.19 %, and 95.26 %, respectively by artificial neural network (ANN). The ANN-MOGA (multi-objective genetic algorithms) predicted higher experimental response values compared to the response surface methodology. A paddy seed germination rate of 94.0 % was achieved using the optimized condition predicted by ANN. Hence, the optimized conditions improved soil and seed distribution precision, ensuring uniform seedling emergence and sustainable paddy production.
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
Copyright 2025 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)