Prototype Platform Perceives Pristine Peppers, Precisely Picks a Perfect Peck

Photo: CROPS Project

Fruit and vegetable harvesting is a task that's ripe for automation. But harnessing the fruits of robotic labor requires more than a succession of terrible puns: there are reasons why we don't have industrial scale robotic harvesting systems yet, especially for high value crops with strict ripeness requirements that are easily damaged by handling.

While several robotics companies have taken intermittent stabs at getting robot harvesters to work on both specific and generalized crops with mixed results, a huge (and well funded) research project from the European Union has recently come to fruition (I'm done now, I swear!) with Clever Robots for Crops: "a highly configurable, modular and clever carrier platform that includes modular parallel manipulators and intelligent tools" to efficiently and reliably harvest fruits and veggies.

The CROPS Project began four years ago and is now nearing completion. Here was the original goal:

CROPS will develop scientific know-how for a highly configurable, modular and clever carrier platform that includes modular parallel manipulators and intelligent tools (sensors, algorithms, sprayers, grippers) that can be easily installed onto the carrier and are capable of adapting to new tasks and conditions. Several technological demonstrators will be developed for high value crops like greenhouse vegetables, fruits in orchards, and grapes for premium wines.

The CROPS robotic platform will be capable of site-specific spraying (targets spray only towards foliage and selective targets) and selective harvesting of fruit (detects the fruit, determines its ripeness, moves towards the fruit, grasps it and softly detaches it). Another objective of CROPS is to develop techniques for reliable detection and classification of obstacles and other objects to enable successful autonomous navigation and operation in plantations and forests. The agricultural and forestry applications share many research areas, primarily regarding sensing and learning capabilities.

A bunch of different prototype platforms are under development, including robots that can autonomously harvest sweet peppers, grapes, and apples, and robots that can execute precision spraying from ground level to canopy. Also on the development list are robots capable of performing generalized tasks like trellising, pruning, leaf picking, and crop sensing and monitoring.

Here is a robot that uses a vision system and an articulated arm to autonomously harvest sweet peppers (the researchers tested two different robotic hands):

There are some compromises that are necessary to get all of this stuff to work. Most notably, the plants themselves need to have their growth arranged to be robot-friendly: apple trees, for example, are carefully pruned to make them almost two dimensional, with very flat, low sides that are easily accessible by robots. You can see some (small) pictures here.

When these robots are deployed in the favorable, structured conditions that they expect, they perform quite well. However, they're not fast, and they're definitely not cheap, and unfortunately, unskilled human labor is going to be much more efficient while it's readily available. Where it seems likely that we'll see systems like these first are in agricultural research, where consistency, repeatability, and sterility are important. Or in greenhouses in space (or on the Moon or Mars), where human labor is at an enormous premium.

[ CROPS Project ]

Advertisement

Automaton

IEEE Spectrum's award-winning robotics blog, featuring news, articles, and videos on robots, humanoids, automation, artificial intelligence, and more.
Contact us:  e.guizzo@ieee.org

Editor
Erico Guizzo
New York, N.Y.
Senior Writer
Evan Ackerman
Berkeley, Calif.
 
Contributor
Jason Falconer
Canada
Contributor
Angelica Lim
Tokyo, Japan
 

Newsletter Sign Up

Sign up for the Automaton newsletter and get biweekly updates about robotics, automation, and AI, all delivered directly to your inbox.

Advertisement