Skip to main content
ARS Home » Southeast Area » Houma, Louisiana » Sugarcane Research » Research » Publications at this Location » Publication #310566
Title: An overhead optical yield monitor for a sugarcane harvester based on two optical distance sensors mounted above the loading elevator

Author
item PRICE, RANDY - Louisiana State University
item Johnson, Richard
item VIATOR, RYAN - Dow Agrosciences

Submitted to: Applied Engineering in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/20/2017
Publication Date: 10/20/2017
Citation: Price, R.R., Johnson, R.M., Viator, R.P. 2017. An overhead optical yield monitor for a sugarcane harvester based on two optical distance sensors mounted above the loading elevator. Applied Engineering in Agriculture. 33:687-693. https://doi.org/10.13031/aea.12191.
DOI: https://doi.org/10.13031/aea.12191

Interpretive Summary: Yield monitoring is a first goal in many precision farming projects. These data can be used to help farmers better manage their fields and, when combined with other farm data such as input costs, can help farmers develop profit and loss maps. An optical yield monitor was designed for use on a billet-type sugarcane harvester. The system uses three lasers mounted on the top of the harvester elevator and measures the cane yield in the field. A calibration equation that was developed with data from the monitor indicated that the amount of cane on the elevator, the harvested distance, the harvester speed, and the direction of cut were all significantly correlated with weight. Cane variety, sampling day, and other sensor reading combinations were not significant. Weigh wagon weights ranging from 0.3 to 2.7 metric tons indicated that the calibration equation using the above parameters predicted weights with a good fit around a 45 degree line, yielding an average error of 1.3% and a standard deviation of 10%. On larger truck loads (over 60 metric tons), the data indicates that the system may be able to predict loads with errors below 2%. The yield monitoring system developed during this study should help sugarcane growers better manage their fields and ultimately increase profits.

Technical Abstract: An optical yield monitor was designed for use on a billet-type sugarcane harvester. The system uses three lasers mounted on the top rails of the elevator and measures both the duty-cycle and height of material on the slats. A multivariate SAS® analysis with data from the monitor indicated that the height of material, duty-cycle, harvested distance, harvester speed, and direction of cut were all significantly correlated with weight, with an overall correlation coefficient of 0.97 for the full model. Cane variety, sampling day, and other sensor-reading combinations were not significant. Weigh wagon weights ranging from 0.3 to 2.7 metric tons indicated that the multivariate equation using the above parameters predicted weights well with a good fit around a 45 degree line, yielding an average error of 1.3% and a standard deviation of 10%. Individual variable analysis were not as good as the multivariate, but still returned favorable results with the duty-cycle and height having linear and polynomial relationships to the weigh wagon weights with R-squares of 0.84 and 0.89, respectively, but average errors were quite high (> 20%). On larger truck loads (over 60 metric tons), the data indicates that the system may be able to predict loads with errors below 2%.