Apples contribute about 21 billion dollars annually to the US economy, yet the orchards that grow them depend on labour that is increasingly scarce and expensive. At one major Washington producer, labour's share of total costs has risen from about 45 per cent fifteen years ago to more than 60 per cent today. A new Cornell-led project aims to change that arithmetic by developing robots for the most labour-intensive orchard tasks: pollinating flowers, thinning fruit, harvesting apples and weeding between rows.
A four-year, 7.5 million dollar grant from the USDA Specialty Crop Research Initiative will establish a Center of Excellence for Orchard Robotics in Cornell's Department of Biological and Environmental Engineering, led by Manoj Karkee. Nine partner organisations join the effort, including Washington State, Oregon State, Penn State, Michigan State, Carnegie Mellon and Stanford, alongside commercial growers and the Israel-based firm Fresh Fruit Robotics.
The technical programme spans soft robotic hands that handle fruit without bruising, neural networks trained to recognise leaves, stems and fruitlets, digital twins of real orchards for horticultural analysis, and autonomous navigation through canopies, trellises and wires. Crucially, the team also includes sociologists and economists studying what makes growers adopt such machines, since single-task robots have historically been hard to justify financially. New York, second only to Washington in apple production, stands to benefit directly.