RoboCrop: Teaching robots how to pick tomatoes

https://phys.org/news/2025-12-robocrop-robots-tomatoes.html

by smurda • 7 months ago

115 69 7 months ago

phys.org

asymmetric• 7 months ago

Tangentially related: https://farm.bot/

GiorgioG• 7 months ago

$5,000-7,000...that's crazy.

daemonologist• 7 months ago

What's crazy is that it only does the easy stuff (planting and watering). What we need is a robot to do the hard stuff (in my home-gamer opinion: pest control and weeding; maybe picking is most relevant for commercial agriculture).

dekhn5• 7 months ago

I understand the desire of academics to help agriculture, but they really need to check in with the field before coming up with prototypes like this, because they are duplicating existing things (ag companies already do this), making hardware that would never survive the field (ag companies already solved this), and obscenely expensive (ag companies come up with better cheaper solutions).

9rx7• 7 months ago

I understand the desire of academics to help agriculture, but they really need to check in with the field before coming up with prototypes like this

Do they? Academia is about the individual. It is not about others. Sometimes what an academic comes up with ends up being applicable to a larger audience, but that's not the goal. Industry is where people try to do things for others.

dekhn5• 7 months ago

I mean, if you're an academic and you don't care if anybody uses your technology, that's fine, I guess, but the folks publishing these papers want the industry to use their ideas: "In the agricultural sector, labor shortages are increasing the need for automated harvesting using robots." is the very first sentence.

esafak• 7 months ago

This is the Engineering- not Literature department. People expect it to be useful.

reachableceo• 7 months ago

Really? Can you link to those solutions please ?

ramy_d• 7 months ago

I'm looking at something like this https://www.youtube.com/watch?v=l4Dc6QNWiIs and I feel like they are doing totally different things. Both harvest tomatoes, but these are totally different approaches.

ruslan• 7 months ago

Did you see tamatoes harvested that way ? They are suitable only for producing sauce.

MisterTea• 7 months ago

making hardware that would never survive the field

What I saw in the article was a prototype and see no reason for it to be "field ready."

asdff3• 7 months ago

It might not survive the field but seems fine for greenhouse setting. And a great way to line up a good computer vision engineering job with an ag firm after.

zzzeek2• 7 months ago

In the agricultural sector, labor shortages are increasing the need for automated harvesting using robots.

This is about Japan, but like the US, Japan has a restrictive immigration policy and an aging, not-replaced population that's at the core of this issue. Japan has been toying with expanding immigration in the area of health care workers [1] recently, but like in the US, there really isn't a labor shortage issue if immigration policy is liberalized.

So this is like so many other things a complex and mediocre technological solution to what's actually a political issue.

[1] https://www.bpb.de/themen/migration-integration/regionalprof...

standardUser2• 7 months ago

I agree about immigration, but the world has a large amount of very fertile land in places with very high costs of living. Bringing in large numbers of new immigrants at ultra low pay will have big consequences in most high-cost countries. It's worked well in the US, but that's because of our (former) identity as a nation of immigrants and the massive overlap between US and Latin American culture. In other nations, the outcome could very well be a racially/culturally incompatible underclass working the lowest paying and least consistent jobs, with little-to-no chance of fully integrating.

zzzeek2• 7 months ago

but Japan is actually relaxing immigration rules due to the need for younger workers. I bet they can pay them pretty well for less than the R&D, maintenance, and loss of productivity of the robots costs

graemep3• 7 months ago

I can understand culturally incompatible, but what on earth is "racially incompatible"?

working the lowest paying and least consistent jobs, with little-to-no chance of fully integrating.

It depends who the immigrants are. If your immigration laws favour highly skilled immigrants that is not going to happen.

In the UK people who live in ethnically mixed areas tend to integrate. In fact, I think most people integrate but the minority who do not are just more noticeable and used politically (not by just one side either).

tracker12• 7 months ago

Some of this is showing in a lot of places already. Cultural adoption as part of migration is important and you can only bring in so many migrants while maintaining anything resembling a national identity. Not to mention secondary effects of bringing a literal under-class of migrant workers into a society already facing the aftermath of heavy inflation combined with wage stagnation.

jjk1663• 7 months ago

Nope, it's a technological issue. Increasing immigration might address some of the symptoms of the issue, but it does nothing to address that a human being still needs to do this labor. Frankly even if you were to liberalize immigration laws, convincing people to upend their lives and move to a high cost of living country where cultural integration is difficult at best just to pick tomatoes is not exactly a trivial task. Even if you do get people to come for menial labor, as you say there are plenty of other areas like healthcare where labor is in high demand, so you're likely still going to be faced with labor shortages in less desirable fields. Immigration is a treatment, automation is a cure.

HanClinto2• 7 months ago

"Tomatoes typically bear fruit in clusters, requiring robots to pick the ripe ones while leaving the rest on the vine, demanding advanced decision-making and control capabilities."

At what point do we begin to grow tomatoes specifically for their harvestability (in addition / as opposed to other attributes)?

This sort of thing happened years ago with farmers producing product specifically for things like "durability in shipping" -- I'm thinking of "machine-pickable" as the natural next step for growers to aim for.

Is this already being done? I'd love to hear about how this sort of thing is already in place.

Whether this means mechanically manipulating flower + fruit locations (specifically growing vines in a way that produces fruit in a controlled manner), or possibly even breeding cultivars that specifically have more robot-friendly fruit clustering, I wonder what these sorts of efforts might look like in the future?

ac29• 7 months ago

The way greenhouse tomatoes are grown is already pretty robot friendly.

See below for a couple examples:

https://www.denso.com/global/en/news/newsroom/2024/20240513-...

https://tta-iso.com/innovations/harvai

mcguirep• 7 months ago

I'm thinking of "machine-pickable" as the natural next step for growers to aim for.
Is this already being done?

This is, indeed, already something that is done. As I understand it, for tomatoes it's typically for canning varieties, but they're called determinate cultivars[1]. Even with those, I know in processing you still have to discard the occasional fruit that isn't ripe.

I imagine this kind of technological solution would also be more useful when picking tomatoes for use as the fresh fruit.

[1] https://en.wikipedia.org/wiki/Determinate_cultivar

Hasz• 7 months ago

It is done for entire species.

There is plenty of fruits (Pawpaw, loquat, soursop come to mind) that are really not grown at-scale commercially in the US due to spoilage, easy to bruise, or other similar issues.

If you like interesting fruit, I highly recommend https://www.youtube.com/@WeirdExplorer/

for many fruits you will have never seen before.

fellowniusmonk• 7 months ago

Loquat cardamom jam is pure sex on a buttered english muffin. Probably the most satisfying flavor I've ever consumed, certainly there is no more satisfying flavor. Sadly just as my tree started producing bumper crops a historic freeze killed it.

phito2• 7 months ago

Will they manage to make mass produced tomatoes when worse than they are now? Seriously they're so bad, they're not even worth purchasing in my opinion.

amitport• 7 months ago

I bet you live in the US. What you describe is not a universal issue.

inemesitaffia• 7 months ago

Most of the issues are from lack of ripeness at plucking.

If you wait till they are about to spoil on the vine, even the one's you really don't like will have taste

anothernewdude• 7 months ago

Keeping them on the vine is far better for the consumer, who can have a range of tomatoes that ripen as you eat them.

iancmceachern3• 7 months ago

"At what point do we begin to grow tomatoes specifically for their harvestability"

This has been happening for hundreds of years already with every food crop.

jjk1663• 7 months ago

Presumably they mean robot harvest-ability as opposed to human harvest-ability.

liamzebedee• 7 months ago

They are doing this now actually for plant breeding! "Engineering crop flower morphology facilitates robotization of cross-pollination and speed breeding" covers one example by breeding flowers to be more easily pickable by robotics.

[1] https://www.cell.com/cell/abstract/S0092-8674(25)00840-2

stuaxo• 7 months ago

We already did this for durability, resulting in tomatoes that didn't taste of very much.

Now, the supermarkets that sold those have solved it by breeding ones that are incredibly sweet.

robotguy• 7 months ago

They're called "processing tomatoes" and it's a very interesting crop and industry. Bred for a narrow ripening window, to be machine harvestable, and shippable in massive bulk.

https://ctga.org/tomato-facts/

ge96• 7 months ago

I'm wanting to do this for dead leaves/pruning

Or if it makes more sense to just let them fall, identify and pick up the leaves from the floor/plant pot

tguvot• 7 months ago

fruit picking robot (trees, tomatoes, berries) few years ago

https://www.youtube.com/watch?v=Uujse8pEvBk

https://www.youtube.com/watch?v=Tp1KtHV9lTA

https://www.youtube.com/watch?v=oxF3Ok6Uf64

https://www.youtube.com/watch?v=wnyfX4LcAfU

metalman• 7 months ago

"Robots will automatically harvest tomatoes that are easy to pick, while humans will handle the more challenging fruits." this is EXACTLY the oposite of the way it is supposed to work, and brings to mind agricultural sayings concerning greed and taking unfair advantage, "taking the low hanging fruit", and "cherry picking" with the blame of "inefficiency" going to fall on the plants, and a bigger push to breed "robot friendly" crops or no doubt "cant we just print this shit?"

sbfeibish• 7 months ago

In my neck of the woods we use deers. And they work at night too.

wingchen• 7 months ago

related company when it comes to agriculture robotics: https://www.beagle-tech.com/

rurban• 7 months ago

Tomatoes are actually easy to pick by robots. The AI part is using the red part of RGB imaging, moving the arm to the fruit and suck it in.

Harder parts are the watering, because that's the part which is always broken. Dutch company.

And detecting diseases. Which is the harder AI part my company is doing.

Moving the heavy robot around is also not that easy, esp. with the cheap dutch robots.

RoboCrop: Teaching robots how to pick tomatoes

Related Stories