Agriculture Reference
In-Depth Information
discriminant analysis (QDA) and
k
-nearest neighbor (kNN)-based algorithms. The
statistical models QDA and kNN yielded high overall classification accuracies of
>80%, with a diseased (HLB) class classification accuracy of >90%.
7.7 CONCLUSIONS
The modern era of fruit and vineyard automation began in the 1950s as producers
sought to improve harvesting labor productivity in numerous crops such as potatoes,
cabbages, and tomatoes. As technologies advanced, development programs for more
complicated production tasks began to emerge using mechanization and automation
concepts. A variety of solutions are commercially available, or currently under devel-
opment for production tasks such as harvesting, pruning, hedging, planting, spraying,
and fertilization. As we have seen in this chapter, most solutions to fruit and vineyard
automation problems are multidisciplinary in nature. Although there have been sig-
nificant technology advances, many scientific challenges remain. Viable solutions will
require engineers, horticultural scientists, plant breeders, entomologists, and patholo-
gists working together who understand crop-specific biological systems and produc-
tion practices, as well as the machinery, robotics, and controls issues associated with
the automated production systems. Clearly focused multidisciplinary teams are needed
to address the full range of commodity specific technical issues involved. Although
there will be common technology components, such as machine vision, robotic manip-
ulation, and vehicle guidance, each application will be specialized, because of the
unique nature of the biological system. However, collaboration and technology sharing
between commodity groups will offer the benefit of leveraged research and develop-
ment funds and reduced overall development time for multiple commodities. We can
be very proud of the progress that has been made in the past 50 years of fruit and vine-
yard automation development, but must also recognize the challenges that lie ahead.
REFERENCES
Annamalai, P., and Lee, W. S. 2003. Citrus yield mapping system using machine vision.
ASAE
Annual Intl. Meeting
, Paper No. 031002.
Annamalai, P., Lee, W. S., and Burks, T. F. 2004. Color vision system for estimating citrus
yield in real-time.
ASAE Annual Intl. Meeting
, Paper No. 043054.
Balerin, S., Bourely, A., and Sevila, F. 1991. Mobile robotics applied to fruit harvesting: the
case of greenhouse tomatoes.
Proc. 1991 Symp. Automated Agriculture for the 21st
Century
, St. Joseph, MI. ASAE. 236-244.
Bedford, R., Ceres, J. L., Pons, A. R., Jimenez, J., Martin, M., and Calderon, L. 1998. Design
and implementation of an aided fruit-harvesting robot (Agribot).
Industrial Robot
,
25(5), 337-346.
BEI. 2009. Available at: http://www.beiintl.com/Rotary_Harvester.html/.
Ben-Tal, Y. 1983. Horticultural aspects of mechanical fruit harvesting.
Proc. of the Intl. Symp.
on Fruit, Nut and Vegetable Harvesting Mechanization
. ASAE. SP-5, 372-375.
Benson, E. R., Reid, J. F., and Zhang, Q. 2001. Machine vision based steering system for
agricultural combines. ASAE, Paper No. 01-1159.
Berenstein, R., Shahar, O. B., Shapiro, A., and Edan, Y. 2010. Grape clusters and foliage detec-
tion algorithms for autonomous selective vineyard sprayer.
Intelligent Service Robotics
,
3(4), 233-243.
