Cornell University is leading a four-year, $7.5 million effort to develop robots capable of pollinating flowers, thinning fruit, harvesting apples and weeding orchard rows as growers face rising labor costs and persistent worker shortages.
The U.S. Department of Agriculture's Specialty Crop Research Initiative grant will establish a Center of Excellence for Orchard Robotics in Cornell's Department of Biological and Environmental Engineering. Nine organizations and researchers from several universities and industry are participating.
Manoj Karkee, a Cornell professor of agriculture and life sciences, is leading the project with Matthew Whiting, a tree-fruit horticulture professor at Washington State University. Karkee said automation could reduce the risks of ladder falls, machinery accidents and repetitive-motion injuries while creating jobs in manufacturing, maintaining and supervising the machines.
Labor costs squeeze apple growers
Labor accounted for about 45% of total costs at Washington Fruit and Produce Co. 15 years ago and now exceeds 60%, technology director Gilbert Plath said. The Yakima, Washington-based company grows apples and cherries across 10,000 acres.
Plath said apple prices have remained near their level of 20 years ago while production costs have risen, leaving growers to absorb the pressure. His family's fourth-generation operation is one of the project's collaborators.
Apples contribute about $21 billion annually to the U.S. economy and are the nation's most-consumed fruit, according to Cornell. New York ranks second behind Washington in overall apple production and first in the diversity of varieties grown.
Cornell Cooperative Extension fruit specialist Mario Miranda Sazo will lead outreach to New York growers. The team will seek feedback to ensure the equipment addresses practical needs and can be adopted at a cost growers can justify.
Artificial intelligence guides orchard work
The researchers want the robots to operate autonomously and drive themselves through orchards. The machines must recognize leaves, stems, branches, trellises, wires and fruit, then decide where it is safe to move and which small fruit to remove so the remaining apples can grow to marketable size.
The project will also create digital replicas of orchards for horticultural analysis and examine the economic and cultural factors that influence whether farms adopt new technology. Participants include specialists in engineering, computer science, horticulture, sociology and economics.
Karkee has worked for more than two decades on soft robotic hands that can handle fruit without bruising it and on autonomous agricultural machines. He said recent advances in neural networks have sharply improved the robots' ability to interpret images and make decisions.
One machine could work across the seasons
Avi Kahani, founder and CEO of Israel-based Fresh Fruit Robotics, said commercial viability remains a central challenge. Growers may not be able to justify an expensive machine that works only during harvest.
The team is therefore pursuing equipment that could prune during winter, pollinate and thin fruit during spring, weed between rows and harvest during fall. Using one robot across several seasons could improve its return on investment.
Academic partners include Cornell, Washington State, Oregon State, Penn State, Michigan State, Carnegie Mellon and Stanford. The work also receives support from federal capacity funds managed by the Cornell University Agricultural Experiment Station and earlier Cornell grants focused on artificial intelligence, food security, sustainability, health and life sciences.
FingerLakes1.com is the region’s leading all-digital news publication. The company was founded in 1998 and has been keeping residents informed for more than two decades. Have a lead? Send it to news@fingerlakes1.com.