Hit enter to search or ESC to close

How Autonomous “Burro Bots” Make Harvesting Grapes Easier for Farm Workers

There’s nothing quite like the plump, juicy deliciousness of a California table grape. But the future of the fruit is uncertain because there aren’t enough workers to perform the strenuous work of harvesting. According to growers, a labor shortage outranks a severe drought as the biggest threat to the crop in California. The solution for increasing productivity with fewer workers might come in an improbable shape: small robotic farm vehicles that utilize sensor fusion, mapping, and machine learning methods reminiscent of an autonomous car.

“Nobody in the U.S. raises their kids anymore to get up at 4:15 in the morning, drive into the field and work in 110 degrees of heat,” says Charlie Anderson, chief executive of Burro.ai, an autonomous farming tech company. The work, Anderson says, is repetitive and arduous and the workers themselves are aging, with the majority in their forties and older. “You take your wheelbarrow, load it up with 200 pounds of fruit, and march it down a row to a pack table. Then, you come back and do it again and again,” he says. He estimates that traipsing back and forth between picking and packing adds up to four or more miles per person per day. 

But Anderson has an alternative. “You can have a robot doing the hauling task all day long,” he said. “Meanwhile, the worker stays in the shade and does the dexterity bit.”

Division of labor

Agricultural tech companies like Burro.ai have invested billions of dollars into robots in recent years. Advanced Farm Technologies, based in Davis, CA, is developing robots to pick berries with a soft grip. UK-based Earth Rover makes field robots to automate scouting and harvesting tasks. FarmWise, with offices in San Francisco and Salinas, CA, designs driverless tractors to weed fields without herbicides. Israel’s FFRobotics and Australia’s Ripe Robotics are developing bots that emulate how the human hand-picks fruit, discriminating between what’s ripe or not. And Strio AI, in Cambridge, Mass., automates labor-intensive tasks for specialty growers, for instance cutting the unwanted runners that branch off of strawberry plants.

These highly differentiated and dexterous tasks help tackle the significant challenge of automating agriculture, beyond large systems for weeding or spraying. Burro.ai’s answer to this is an autonomous cargo-carrying platform (also called Burro) for a specific use case: alleviating the need to carry 40 or 50 pounds of grapes from the vines to the packing station. Anderson wisely selected to start with a task offering the greatest immediate benefit to the human workers themselves.

How it works

We spoke with Anderson on a hot August morning. He was in the field in Mettler, CA, during a typical day of harvesting table grapes. At 5:15 in the morning, a trailer with about eight four-wheeled Burros—similar in shape to a pickup truck bed (minus the cabin and engine bay)—rolls into the field. A few minutes later, the crews show up. 

Each team of six to eight workers grabs a Burro off the trailer, turns it on, and pushes the “Follow” button. The robot then uses its A.I. vision system to lock into one or more people in that crew. Thanks to a global positioning system (GPS), a powerful graphical processing unit (GPU), four inertial measurement units (IMUs), a 4G modem, and four cameras, the robotic platform positions itself to alleviate the task of lugging heavy bundles of fruit from the vine to the packing stations. That sounds simple enough, but Anderson admits that “quite a bit of special sauce” goes into the sensor fusion and algorithms. GPS is not always reliable in remote fields, so computer vision is critical.

Burro.ai's Burro bot navigates fields of grape vines

Much as self-driving cars develop sophisticated approaches to challenging roadway conditions over time, the machines learn as they go. Sometimes, the Burro needs to drive over weeds and other obstacles. “At the same time, if it sees a vine, it better stop for that thing to avoid the risk of damaging the crop or the fruit,” says Anderson.

Simplifying the worker-machine interface

Robots need to disengage when the course of action is too challenging. Crew members are trained to clear work stoppages by pushing a button twice. Each intervention triggers imagery to be recorded and automatically sent to Burro.ai’s technology offices and command center in Philadelphia. “We track autonomous miles between interventions. And as that number rises, we’re able to pull our engineers out of the field more and more.”

Locomotion is provided via four wheels independently driven by individual motors and wheel encoders for precise odometry. Safety is paramount. These 400-pound robots move at a leisurely pace of a meter per second. There are warning lights and sounds—and soft bumpers front and back. “We do localization, mapping, perception, and a pretty intense fusion of GPS and vision that has taken a lot of time to perfect,” said Anderson.

Rodolfo Medina, 64, has been an agricultural laborer in California’s Central Valley since 1973. For nearly 50 years, his primary tool has been a wheelbarrow. The Burro bot has changed his work life. “I’ve never seen anything like this out here,” he said. 

Burro.ai's Burro bot helping farmers harvest grapes

Mayra Medina, Rodolfo’s daughter, about 30 years his junior, works alongside her father. “Some things were complicated at first, but after a few days, you get used to it,” she said. 

“There are a lot of benefits because you don’t get tired anymore, and you have more energy for other things.”

While Burro.ai is focused on the immediate use case of hauling heavy loads, the company will slowly expand the capabilities of its robotic platform. For example, its vision system can handle such delicate tasks as counting grapes or disease detection. 

There’s hard work ahead—especially as dexterity is added to the machines. said Anderson. “You’ve got to build hardware and software, and integrate it … hammering out edge cases to get something working,” said Anderson. “Autonomous systems [for agriculture] are like the Mount Everest of hard businesses to solve.”

Choose your lane

How Autonomous Vehicles Distinguish Between Bicycles and People Who Ride Them

How Autonomous Vehicles Distinguish Between Bikes and People

When it comes to how autonomous vehicles see the world, humans come first, literally. Autonomous vehicles (AVs), like the kind operated by Pittsburgh-based Argo AI, use Machine Learning to detect and classify the objects in their surroundings, identifying people...
Why The League of American Bicyclists is optimistic about autonomous vehicles

Why a Leading Cycling Advocacy Group Is Optimistic About Autonomous Vehicles

As autonomous vehicle use grows, AV companies and the League of American Bicyclists are collaborating on how to ensure cyclists and motorists can share the roads safely, even if the “motorist” is artificial intelligence software. As part of the...

Self-Driving Is Arriving Right On Time. Just Like Ice Cream Did

Seven years ago, I was a self-driving skeptic. Not of the technology. Of all the “experts” promising autonomous vehicles would be everywhere by 2020. You didn’t need to be Nostradamus to know that was ridiculous. All you needed was...
Illustration of a futuristic parking deck turned into a mixed-use space, with AVs driving by

How Autonomous Vehicles Could Help Transform Parking Lots

Researchers say it’s likely that autonomous vehicles (AVs) can help reduce the need for parking lots, opening more room for grass and trees and other elements of nature. It may not seem like it when you’re circling the block...
An illustration of an Argo autonomous vehicle in teal against a backdrop of buildings, a bicyclist, and research papers

7 Big Breakthroughs From Argo Research at CVPR 2022

The 2022 Conference on Computer Vision and Pattern Recognition (CVPR 2022) is nearly here. Thousands of computer scientists, software engineers, and researchers from around the globe will gather in New Orleans to review and discuss their latest work in...

Researchers Predict the Future With Lidar Data

Researchers at Carnegie Mellon University’s Argo AI Center for Autonomous Vehicle Research, a private-public partnership funded by Argo for advancing the autonomous-vehicle (AV) field, say they have come up with a way to use lidar data to visualize not...

Must Reads