|Upadhyaya, Shrini - UNIV OF CALIFORNIA, DAVIS|
Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 1, 1999
Publication Date: September 13, 1999
Citation: Pelletier, M.G., Upadhyaya, Shrini K. Development of a tomato load/yield monitor. Computers and Electronics in Agriculture. 1999. v. 23. p. 103-117. Interpretive Summary: The development of yield monitors has led to a revolution by which growers can now tailor each part of a field to it's own specific requirements. This is now known by the term precision agriculture or farming by the foot. Currently there are only a few crops where suitable yield monitoring systems have been developed. One of the leading revenue crops for California is processing tomatoes. Before this study there was no available system for measuring crop yield as the crop is harvested. Given this, the objective of this study focused on the development of a suitable yield monitor for use in processing tomatoes. The yield monitoring system that was developed consisted of a very accurate ground based differentially corrected satellite Global Positioning System (DGPS) that was monitored by an on-board computer that also monitored the mass flow of the fruit as it exited the harvester and was loaded onto chase transport gondolas. During the testing of the system, it was shown to perform at a level of accuracy yielding a coefficient of determination equal to 0.999. The results also detailed that the level of crop non-uniformity was in excess of +/- 20 percent of the average yield. This clearly demonstrates the remarkable gains that can be achieved through the pursuit of precision agriculture.
Technical Abstract: The objective of this study was to develop a load/yield monitor for mounting on a processing tomato harvester. A continuous mass flow-type yield-sensing device equipped with load cells and a vibration-resistant angle transducer, along with a differential global positioning system (DGPS), was developed to collect spatially variable yield data in real- time. Both analog and digital signal processing were employed to maximize the accuracy of data acquisition. The yield monitoring system was calibrated, validated, and tested during the 1997 harvesting season. The coefficient of determination, r2, between the actual fruit weight and the yield monitor prediction, was 0.997. The system was then used to measure yield variability and produce yield maps. There were significant spatial variations in processing tomato yield. The lowest 20percent yielding area within a field produced less than half the tomatoes compared to the highest t20 percent yielding area within the same field.