Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Farming robot kills 100k weeds per hour with lasers

freethink.com

Post reply on HN
Print view View original

171–180 of 483 posts

Re: Farming robot kills 100k weeds per hour with lasers

#171

Post by ashtonbaker »

janglytim wrote: post #122
ashtonbaker wrote: post #14

Earlier quoted context omitted.

I'm all about regenerative agriculture, permaculture, etc. But I've never seen a solution to weeds other than to spend a lot of manual labor removing them. Is there a method that reduces the number of weeds that appear, or reduces the need to remove them? I tend to think that solutions like this offer the possibility of the best of both worlds - automated (and therefore scalable) agriculture without the chemical depe…

What about using animals to help with weed control[1], for example ducks used in rice paddies[2] or weeder geese? Also mulching and growing cover crops alongside and in succession with our crops [3] prevents weeds without as much labor as traditional wedding. For example planting clover around crops, which stays short and fixes nitrogen while competing with weeds. I wish we would spend more time rethinking our indust…

The link at [1] seems to be mostly about "pasture and rangeland weed control" with limited applications to forestry and crops. The examples for those are good, but seem at a glance application-specific. The most general recommendation for crops is to use animals during the pasture rotation of a plot to reduce weed pressure during a growing rotation.

[2]: Again, great, but very crop-specific.

[3]: I've seen/used mulching on permaculture farms, and it seems like a fairly effective method. Also has the benefits of reducing the need for water, and regenerating topsoil via decomposition. But labor-intensive. The clover idea is very interesting and new to me.

> I wish we would spend more time rethinking our industrial farming practices, rather than try to prop them up with diesel burning robots.

Agree wholeheartedly! And if the answer is that more labor is required to farm sustainably, then I'm personally all for that. But I'm also pragmatic, and if we could improve the economics of sustainable farming by automating some of the manual labor, I think that would be great. I'm not sure this specific robot addresses that - it seems designed for industrial monoculture farms - but it's an interesting idea to me.

Re: Farming robot kills 100k weeds per hour with lasers

#172

Post by notfromhere »

pitaj wrote: post #149
danbruc wrote: post #109

Earlier quoted context omitted.

No, the price increase will be much smaller than the wage increase because the wage increase only goes to the fraction of the population that does the work while the price increase applies to the entire population.

If you raise the base cost of production in a low margin case like agriculture, then the price of the good must increase or they would be running at a loss.

Agriculture is incredibly subsidized, which means cost of production and cost at point of sale do not have a tight relationship

Re: Farming robot kills 100k weeds per hour with lasers

#173

Post by SamBam »

the_mitsuhiko wrote: post #44
cmrdporcupine wrote: post #20

I want to see details on power source and energy usage. That's where I'm a bit skeptical. To do the 20 acres they're talking about you'd need one hell of a huge battery, assuming it's electric. Even just to move the thing. Then add power draw for the laser on as well. Also seems odd to have it be something that is self-powered and autonomous rather than just something you pull behind a tractor on the three point hitc…

Farming equipment is typically fuel based. And this appears to be the case for this robot as well: it uses Diesel.

At that point, is this more environmental than applying Round-Up to the field?

Each of these lumbering beasts goes 5 MPH, burning diesel the entire time. How many gallons of gas is going to be spent weeding a hundred acres of farmland?

Re: Farming robot kills 100k weeds per hour with lasers

#174

Post by goodpoint »

MayeulC wrote: post #125
motohagiography wrote: post #56

Given where ML and CV are today, I'd bet on open source models trained on weeds within 3-5 years. This company can scale, but their unit price is going to plummet. Also, it sounds like they're using cameras, where it's concievable that other future sensors could be more efficient. An iteration of this with higher resolution cameras and small flying drones seems like an intuitive next step. This is a super interesting…

I don't think you really need AI/CV if you have multispectral or full-spectrum imaging. Different species will have different absorption/emission at multiple wavelength. This makes it easy to identify crops, and seems to be commonly used for aerial imaging[1-3]. That kind of sensor is expensive though, and while you could probably do it for cheap with something like a DLP wheel (edit: or an array of different light e…

> you have multispectral or full-spectrum imaging

...that's exactly what CV needs to handle.

Re: Farming robot kills 100k weeds per hour with lasers

#176

Post by bluGill »

JumpCrisscross wrote: post #117
sprainedankles wrote: post #108

Earlier quoted context omitted.

I'd push back on that a little bit. I grew up on a farm that's still operating today, and it truly is difficult to find people to work those jobs. The farm pays just as well as any other job in my rural hometown, so from my experience, it seems to be more of an issue with the work itself. The hours are longer, and work is more sporadic/seasonal. When harvest rolls around, farmers need to get the crop out of the groun…

> pays just as well as any other job in my rural hometown, so from my experience, it seems to be more of an issue with the work itself. The hours are longer, and work is more sporadic/seasonal If it pays as much as other jobs with shorter, less sporadic hours, it's underpaid.

Over the course of the year you make the same. However some months you make a lot more/less than others.

Re: Farming robot kills 100k weeds per hour with lasers

#177

Post by MayeulC »

riskable wrote: post #147
MayeulC wrote: post #125

Earlier quoted context omitted.

I don't think you really need AI/CV if you have multispectral or full-spectrum imaging. Different species will have different absorption/emission at multiple wavelength. This makes it easy to identify crops, and seems to be commonly used for aerial imaging[1-3]. That kind of sensor is expensive though, and while you could probably do it for cheap with something like a DLP wheel (edit: or an array of different light e…

> This makes it easy to identify crops ...until they evolve to emit the same spectrum of light as the crops being harvested! Cuz you know that's going to happen.

Well, that's right, and I didn't think of it. But besides what the sibling comments pointed out, that's an issue with any kind of weeding, including manual.

The answer is to select according to a wide range of criteria, and not rely on a single one. That way, weeds cannot progressively acquire resistance, and need to check all the boxes at once, which makes it highly unlikely that they will pass on their "slightly better" genes.

And of course, the larger the scale you employ a single weeding system at, the more risky it gets. It would be great for supplementing herbicide (while lowering doses) at first, for instance.

Re: Farming robot kills 100k weeds per hour with lasers

#178

Post by partiallypro »

joosters wrote: post #38

"It's harder to find people to do that work every single year" Translation: The wages we pay are crap and we're not going to pay them more.

They actually pay pretty well considering ($15-20/hr in some cases,) it's just very hard work. You also have to consider that farms don't make much money. They don't have massive margins like Google and a single bad weather event can wipe out any profit for the year.

Re: Farming robot kills 100k weeds per hour with lasers

#179

Post by MengerSponge »

Balgair wrote: post #131
motohagiography wrote: post #56

Given where ML and CV are today, I'd bet on open source models trained on weeds within 3-5 years. This company can scale, but their unit price is going to plummet. Also, it sounds like they're using cameras, where it's concievable that other future sensors could be more efficient. An iteration of this with higher resolution cameras and small flying drones seems like an intuitive next step. This is a super interesting…

> Also, it sounds like they're using cameras, where it's concievable that other future sensors could be more efficient. Bit of a pedantic note: What other sensor could you use? I'm grasping at straws (pun not intended) to figure out any other modality that could work as well in a giant field of psuedo-randomly mixed plants with the wind blowing chemical signatures all about.

I assume they mean "visible spectrum" when they say cameras, because that's what most off-the-shelf systems are tuned to.

A spectrometer is a single pixel camera, I guess, but it isn't being used with imaging optics, and it isn't being used to stitch together a photograph.

In a general case you could embed some version of GFP instead of glyphosate resistance. Then you can set the system loose (within the field, lol) to actively interrogate plants, zapping intruders that fail their scans.

Re: Farming robot kills 100k weeds per hour with lasers

#180

Post by DennisP »

swiley wrote: post #70
motohagiography wrote: post #56

Given where ML and CV are today, I'd bet on open source models trained on weeds within 3-5 years. This company can scale, but their unit price is going to plummet. Also, it sounds like they're using cameras, where it's concievable that other future sensors could be more efficient. An iteration of this with higher resolution cameras and small flying drones seems like an intuitive next step. This is a super interesting…

Commodity drown with high powered laser for killing things based on image recognition sounds like sci-fi dystopia.

True, but it seems less dystopian than soaking vast fields with millions of pounds of neurotoxins, like we do today.
Post reply on HN