Exhibiting as a tech startup on the Fruit Logistica expo floor, the Israeli-based company has developed a machine with 12 robotic arms, with six pickers on each side of the machine, according to Kahani. Called the FFRobot, the machine can be adjusted based on the width of the orchard row.
“This (past) year we tested in an apple orchard in November and December,” he said.
The FF Robotics machine enters the apple row and then automatically puts down stabilizers and picks the area on each side, picking fruit based on size and color. The machine requires one human supervisor, Kahani said. After picking the section, it automatically moves to the next part of the orchard.
The machine has been under development for nearly five years, the co-founders said.
Perhaps the biggest challenge early in the development of the machine was attracting grower interest.
That has changed, Kahani said.
With average wages for harvest labor at an average of about $15 per hour in the U.S. and Europe, and rising, he said growers from all over the world are interested in cutting labor needs.
“A few years ago, growers didn’t want to hear about mechanically picking fruit, but over the last two years, we are hearing from growers everywhere — including China — that they have a shortage of labor,” he said.
Kober said the company plans to sell its first five to seven machines this fall to early adopters and then go into heavier production by the beginning of 2020.
“We are in good contact with large growers in Washington state, and they are just waiting for us to come with a machine (for sale),” Kober said.
The machine does a good job of handling fruit, with tests showing 3% to 5% damage to the fruit, compared with 10% to 15% damage common from human pickers, he said.
Some of the toughest engineering problems was programming the robotic arms to work together, Kober said.
“The decision to send this arm to this place, that arm come to that place (was difficult),” he said.
“And if you have a couple of apples like twins, if you pick only one, the other one falls down,” Kober said. “So you have to design the system so that when there are couples, you just go together with two arms and you pick them simultaneously so no one falls down.”
Kober said the company’s calculations — based on grower input — indicates the machine could provide a return on investment in two to three years. The cost of the machine is expected to run from $300,000 to $350,000,