As a hobby gardener and wannabe roboticist I've thought a lot about this problem. Before I continue I love this thing and am seriously thinking about getting one.
However, a concern that I'm sure they've thought a lot about is in it's technique. I really only care about making sure there are no weeds within, say, 8 inches of my plants (or put another way, my level of concern about a weed is proportional to it's distance from my vegetable or flower plant). This looks like it will weed everything except the 8 inches around the plant. Never mind things like cucumbers that spread out on the ground and have important bits that are short and can't be contained by the wire protectors. Still, an exciting place to start.
I do raised beds as well, so I have the same concern. But from my point of view the larger issue is that I'm almost never only weeding. Like, it's just not that big of a job.
But more importantly, time spent "in the garden" includes a lot of inspection and minor adjustments to all kinds of things— move plants on a trellis, pinching suckers off the tomatoes, clipping blighted leaves off of plants, ensuring flowers are getting fertilized, monitoring fruit growth, etc.
All of this stuff is an essential part of being in tune with what's going on out there, and I'm not sure that having a robot do one small piece of it is beneficial. Basically, this product doesn't really look like it was built by a gardener— it looks like it was someone who had a solution they wanted to apply elsewhere, and spent 15 minutes watching a gardener at work, without really understanding what they were doing.
In my opinion, that suffers from the same fundamental issue that it's covering a small portion of the visible work that a gardener is doing, but isn't accounting for all the observation and tweaks.
Basically I'm just not sure it's realistic to pitch this product to someone who wants to have an outdoor garden but spend little to no time actually tending it. I feel this kind of user might be better served by something like Aerogarden.
In any case, it's a much harder problem because watering even a small garden takes a _lot_ of water, so for it to be practical, you either need a robot with a huge (and very heavy) reservoir, or a base station at which it can fill up (which would almost certainly require leaving the garden bed itself, massively complicating the whole system).
I agree (as someone who spends all my time in my garden when I'm not at work). I enjoy being outside, getting fresh air, clearing my head, and tending to my own garden. There is something really refreshing about doing it and I'd have zero interest in automating away any portion of it. Just my 2 cents...
I brew up and spray compost tea in 5 gallon batches. A machine that would continuously brew tea and go walk around the garden spraying it everywhere would be pretty cool. But also far more of a niche.
what might be cool is a method by which the robot could tell you the garden needs to be watered. so some sort of sampler to make this determination. get really advanced with optics and we could eventually have it determine if there is a bug problem or disease.
the robot doesn't necessarily need to do all the work, just identify what needs to be done beyond its specialty
It would have to be a much stronger robot to haul around a garden hose. Those things can easily weigh several kilograms when they're full of water, not to mention they can get caught on obstructions rather easily.
There's always a sock waiting to be stuck on somewhere...
But I don't get it why they won't make the robot reverse a short bit and try to go around instead of just rolling over and die.
That would solve 99% of the situations I have rescued my old Roomba and current Neato from. It's very seldom they actually got something entangled in their brushes.
It already is damaging things as it is. The Neato have peeled a USB cable and an audio cable. It also have chewed on a few laptop charger cables but they seems to be tougher so no damage. Hopefully it won't peel something that could be dangerous.
I doubt it could be that hard to detect an unusually low rotation speed. Now it seems to chug until it really stalls. Instead it should stop the brushes immediately and retreat, then retry with just the fan running.
I'd settle for a garden hose that doesn't perish from UV damage after a season and doesn't kink every 7 seconds. Sounds like a pretty simple thing to get - it isn't.
Things would get tangled in a hurry with the hose getting caught up or stuck. Low pressure would help but I still don't think it's viable on anything but row based gardens.
My family kept a large garden for the duration of my childhood. I was made to weed it. It was absolutely that big of a job. It took a couple hours. Every day.
It looks like they started with a Roomba and went from there, which leaves them with a robot that can't deal with raised beds or (as you said) weeds integrated with plants. I have raised beds and then raised side gardens which would render this device useless to me. Same with the other hobby gardeners I know.
I always imagined something with long spindly legs that would cut, inject or spray things that it identified as weeds by leaves. For me, the weeds that are the most trouble are those close to or mixed in with plants. Anything on its own isn't that hard to pull out.
Roombas have to deal with stairs while this thing will probably get it's underside covered in mud (especially since it has that unshielded trimmer) and is supposed to run only inside a fenced area, so I doubt it has any provisions for that.
Easiest way to do it is to put layers of wet newspaper then black mulch. I left it for a month and it's as if the weeds were never there. I made holes in the mulch and planted some veges, they are doing great (except for the Shiso which something ate).
You do know you can buy weed barrier fabric and cover the beds. When you plant you put holes in the barrier where your plant goes. Work well for a few people I know and it's similar price to mulching every season.
Mulch worked for me but I am using pots at the moment.
My garden was literally only weeds (our house is my wife's grandparents and the house had sat mainly abandoned for about 3 years when we moved in). After doing the newspaper and mulch technique and leaving it for about 6 weeks the weeds had all decomposed into soil and haven't come back.
The bit that shouldn't surprise me but still does, is the volume change. Add 6 ish cubic metres of mulch and 5 years later is a thin layer of soil comprising maybe 1-2m.
I expect that one thing that will come out of deep learning is mechanical weeding for industrial level agriculture. Instead of spraying crops with herbicide, mechanically remove weeds. Bonus is agricultural machinery tends to be high capital cost anyways, which allows for a lot more processing power and sensors than a garden robot.
I really hope we get there soon. Not only could this eliminate usage of herbicides, but many smaller autonomous robots vs one really big John Deere could also remove the need for monocultures. And that would really be a major improvement for both crop yields and the environment
Except although the movie does say battery. That's not the original plot device that was written. They were originally written as human based cpu computers. But some higher up thought people wouldn't understand "human processors" so just everytime you think about or watch the matrix again. Just think it's a poor mistranslation that made it into the movie, "processor->battery"
I'm paraphrasing, but Neo says "that's impossible (in reference to the Matrix) since it violates the laws of thermodynamics"
To which Morpheus replies "and where did you learn about the laws of thermodynamics?"
If you're in the Matrix, your knowledge of the Matrix would come from the Matrix itself which would have a bias towards falsifying information to preserve itself.
I remember reading Indians of the Northeast planted corn, beans, and melons together. So switching to polycuture crops should have advantages.
One other thing I've read is that annual plants were the easiest plants to domesticate because the short life cycle results in faster evolution. However perennials are likely way less environmentally destructive.
> What will all these people do once farmers are unnecessary?
There aren't that many people working on farms in developed countries. In the US, only around 3 million people (or <1% of the total population) work on a farm. [0]
In other countries where many people are involved in agriculture, the pending industrialisation/automation of farming will probably have a similar impact to what happened in North America and Europe over the 20th century: more people will move to urban areas and add value to the economy in other ways.
>What will all these people do once farmers are unnecessary?
The answer is "whatever they want". The big problem facing governments everywhere right now is how to jump over the little gully of poverty that comes right after the change to get there.
corn is a tall crop, soy beans are short. Farmers are already experimenting with things like 4 rows of corn 4 soybeans. The corn rows shade the soybean rows and so reduce yield - but in turn the soybeans are not shading the corn so they get slightly better corn yields. This is still experimental, but so far it looks promising.
The best part? Weeds will still evolve resistance — just like they do for herbicides — but they'll do it by growing into shapes that the ML models misclassify.
Updating the robots can be done much faster than it takes for the weeds to mutate. And for mutations to happen, some proportion of the weed population must survive in the environment, as no weeds can not mutate. I doubt they have a chance, unless some species figure out how to live under the earth, in which case, we won't care as those would not be "weeds" any more.
1) Getting very close to the crops and intertwining with them to make weeding very hard
2) Looking almost exactly like real crops (as-in, to the point where most humans can't tell the difference and the machines need to start taking genetic samples)
You're totally right though; the machines have an adaption-speed advantage, so I'd expect this to be a very effective technique overall (assuming that the robotics pan-out).
Until the weeds become actually identical to the actual crops, and we just have weirdly recreated the crops. This is hilarious, and actually precisely how Generative Adversarial Networks work (https://www.wikiwand.com/en/Generative_adversarial_networks), although obviously with gradient descent in the generator network, not biological evolution
Give it a few decades and the weeds will find a way of corroding robots. In a century, we'll have to develop robots that can handle the acid that plants will be spraying on them.
Yep. John Deere will collect all of the recognition data and not allow their customers to see anything except high-level statistics (for an additional fee). Then they'll use it to train next year's model and charge for the upgrade.
Yet dandelions have genetically modified for lawn mowers, growing strangely mutated so as to be flat on the ground, yet the mowers haven't changed. So, sometimes the technology is not as fast as we think.
My source is simple observation. I did all the mowing as a kid, acres every week. In the spring, chopping down all the new dandelions was satisfying and the grass looked great. By the middle of the summer, dandelions were back but now they were strangely 'smashed' ones that grew no taller than 4", with flattened stems that turned or curled or flowers on short, short stems that bent over.
I'm sure that gene was always there; but billions of mowings over a century of lawn care means its present everywhere. Simple selection?
Ahh, no this is similar to the process that causes Bonsai to work. Dandelions have a fairly massive root with plenty of reserves to regrow after something eat's the top. Mowing simply causes similar responses as a goat would.
Plants can't move around so they evolved to deal with a really wide range of environments and stuff eating them etc.
Nah. Dandelions don't regrow stems; this can't be something done in response to being mowed. The plant doesn't think "I got mowed! I guess I'll grow a different way now".
Its a different growth pattern in the plant. Gotta be driven by genetics.
Dandelions are actually perennials so they will regrow stems several times, and survive multiple years. They will also tend to flower most abundantly in spring, but can re-flower in the fall.
Further, it's not just the stem involved, if your mowing regularly your taking out the leaves.
Yup the flat ones were none of those. They were really weird knarly mutated plants with exactly the same structure, leaves, stem and flowers as dandelion but grown flat and warped to hug the ground.
No! And it needs to be done! I've lived in only a couple of places, and the genetic variety of dandelion (dent-de-lion?) is marvelous. When I grew up they were small modestly sized plants. Then the mowing thing, and a 2nd form of flat plants. Moved to California, they were 4' tall with greyish-green dusty leaves. Moved to an acreage in Iowa, all over the fallow field one year they were bushes about 18" high with 100 blooms - you could have sold them in a garden shop for landscaping!
Mendel should have used dandelions for his gene studies, there is lots more to see!
Meh, there are plenty more cracks in sidewalks, open fields, and poorly kept yards than there are mowed lawns. Dandelion evolution probably isn't optimizing for mowed lawns any time soon.
I find it difficult to believe that any optically-driven machine-learning weeding robot will be 100% successful at eliminating every weed in a garden. There will be some overlooked and they will mutate.
Are you forgetting that you're talking about two different processes that operate on orders-of-magnitude different durations?
By the time any evolutionary affects are seen, if at all, of weeds being pulled by ML, ML will have progressed far beyond where it is now, likely well beyond what could be fooled by simple diversity.
DL models can learn to classify hundreds of weed classes in less than a week. It takes centuries for a new breed/trait to appear. This is a very uneven race.
Another use case is invasive species control in parks and large public lands. As an example, Portland has a large park that stretches along a ridge-line for about ten miles, and it's getting overgrown by English ivy, which kills the trees. It's a big enough area that pulling it all by hand would be very expensive. Ivy-pulling robots would be pretty awesome, but it's a hard problem. Vegetation is thick and the terrain is everything but flat. You'd probably need some giant hexapods or something just to move around, and maybe some kind of Ivy-grabbing appendage.
Just mapping out the ivy infestation with drones and image recognition software would be impressive. There's enough room for a quad-rotor to move around in theory, but it would have to have some pretty sophisticated obstacle avoidance algorithms.
Forest park is about ten square miles, and maybe a third is overrun by ivy. That's solvable by a sufficiently large human work crew, but it wouldn't be easy.
Developing, manufacturing, and operating a crew of robots just for this one particular ivy problem wouldn't be worth the cost, but invasive species are a widespread problem and are sometimes on much larger scales, like hundreds or thousands of square miles. At some point, using people isn't a realistic option. If we can solve larger invasive species problems for less money than a human work crew, then that seems like a win.
I envision a robot gardener that would create a paradise from a plot of empty land. Each plant would be placed, cared for and trimmed in such a way as to obtain an artistic effect.
Another kind of bot I am hoping to see is a waste collector. They would find plastic bottles, bags, glass and iron in forests, on roadsides, and even underwater, around beaches. Unleash 100 bots in a forest and in one year, it's pristine. We can also have some collect floating plastic from the oceans.
A trash bot would really be something to see. I'd pay a lot in taxes for that. Add in some ML matrix-stirring to watch out for fauna, and you have taken out ~70% of forest rangers jobs. Heck, just make a Park Ranger bot, cleans trash, scolds tourists, takes entry fees, fights fires; it'd be pretty nice.
That depends. Farmers used to mechanically remove weeds with a cultivator. (a fancy hoe attached to a tractor) Those cultivators will remove all weeds and no weed grows resistant to them.
However they have many disadvantages:
Cost - $500 worth of fuel for the tractor vs $250 for chemical + $50 for fuel to run the sprayer is the big one (These cost numbers are what I've heard but I might remember wrong).
CO2: that $500 worth of fuel vs $50 directly translates into CO2, which in turns translates into global warming. (this is the only one that you can argue doesn't have a financial impact on the farmer - and it isn't clear that those arguments are valid)
Soil disturbance. Every time you til the soil you disturb the natural soil. It isn't as bad as deep plowing, but it is still harmful.
Soil compaction. A repetitively light sprayer can cover 100+ feet of crop in just two tire tracks (the first wheel the runs over the ground does the damage, the second barely adds anything so the 4 wheels result in two tracks). A tractor is heavier (the impact of weight it not linear though so figure 10% worse), and can only do 30 feet so you end up with just over 3x as many tire tracks compacting and harming the ground.
Labor cost. Sprayers can run at 15mph in the field. Cultivators are probably running at 5mph. Not only are you taking 3x as many passes, each pass takes 3x as long. You pay the operator by the hour.
Now for your robot which applies? The weed pulling robot will only disturb the soil where the weed is. However the robot needs to get to the weed which increases compaction along each row.
My guess is farmers will spray everything as a primary prevention. Then they will fly drones over the field and send a robot down only the rows where there are weeds. That is just a guess though, you can imagine your own parameters for future robots which will change everything.
Note, herbicides are not nearly as dangerous as popular press makes them out to be.
When farmers plough they tow the unit behind the tractor, and this removes the compaction of their tyres.
You have to plough each year before sowing anyway so I'm not sure it's a simple as spray v plough.
You must not be familiar with modern farming. Most farmers do no till these days. They never plow - ever. The first seven years of no till they do see less yields than their counterparts that plow, but after that the advantage of non-distributed soil mean they get better yields. (more profit ever before we account for the fact that plowing uses a lot of expensive fuel)
Nothing is simple, but there is a reason most farms and all the big ones are no or limited till.
No. Large size monoculture should rely on preventive methods to control weeds, such as always covering the soil. It's even possible to plant seeds on an existing live cover.
No plowing, no pesticides, just preventive coverage and good tools.
In the long run, those large monocultures don't make sense anyway.
Robots can be great for a lot of high maintenance, Human intensive, high value cultures such as fruits, greens ...
I guess they could also help expand vegetables farms to non cultivated areas, in some autonomous ways.
Allow me for larger anticipation. Light and tiny electromechanical machines (aka cyborg insects) that can be released to monitor and clean weed, other insects, maybe bacterias.
Genetically modified squirrels! They pull weeds, put them in a bin and get a kibble! Its work or die, which keeps the gene line pure. And self-replicating too.
I don't really understand what he's talking about. Most farm equipment nowadays has thousands in dollars of GNSS equipment and a multitude of other sensors.
Because they will probably be purchased by the thousand, if not 10 thousand. Furthermore, they will be exposed to rugged environments where every extra component that can fail is a huge liability.
Exactly. Better to figure out a novel way to use fewer sensors equally well (cheaper) or the same sensors to better effect (more results).
Those things are hard though, plugging a few sensors into a Pi, slapping a LAMP stack on it then creating a web interface and smart-phone app you can show the VCs is easy by comparison.
slapping a LAMP stack on it then creating a web interface
You know, that'd be really great and all, but please, please, please don't discount the potential for rampant, malicious destruction of food supply for kicks, revenge or maybe even strategy/tactics in state conflict. Consider how warlords and pirates operate in other countries and failed states.
Food security is important, and a robot that can selectively clip and gouge weeds can do the same to both critical food and cash crops too.
Furthermore, simple competition, big corporate espionage and personal differences can still affect small organic growers as well. Human factors are real, and novelty is good but it doesn't cure quite everything.
This seems an important point that hadn't occurred to me. Our food chain is already pretty dependent on Internet-connected infrastructure working correctly, but it seems absolutely plausible that this will be a future attack vector if we network and automate farming machinery.
The threat is not from unknown robots sneaking in to uproot crops (after all why not do something fast and cheap like spreading salt?). The threat is that the farmer's own weeding robots will be hacked to uproot crops, probably by ze Russians. Firmware updates should be secured.
Actually I like that, it's closer to the essential truth.
I think George Carlin nailed it with his routine about flamethrowers.
> The very existence of flamethrowers proves that sometime, somewhere, someone said to themselves, 'You know, I want to set those people over there on fire, but I’m just not close enough to get the job done.
As a Roomba owner, I have to say that I'm not really impressed with their decision to make another product when their first one barely even works.
> "Some 15 million of them are cleaning floors all over the planet, and they’re doing so reliably and affordably and autonomously enough that people keep on buying them"
It is absolutely not "reliable". It's barely better than cleaning the floor myself. Some days, cleaning the floor myself would actually be easier, or ends up being what I have to do anyway. The number of issues I've run into with the damned thing is absurd and I feel like they should be fixing those before making a new product.
I think people are for the most part getting tricked into buying these. They're not that great.
I’ve had a Roomba 780 cleaning my apartment daily since 2013. Issues I’ve encountered so far:
1. It occasionally doesn’t properly return to the home base, because it’s hard to find a good place for the home base where it can’t be moved/knocked over accidentally. This happens maybe once every 2 or 3 weeks, and is just a mild annoyance: you just need to carry the Roomba to its home base yourself.
2. After 3 years of usage, the battery died in such a way that it emptied rapidly enough for the Roomba to not be able to return to the home base. This was easily fixed by installing a new battery.
I recently upgraded to the Roomba 980, and it seems to work better than the 780 — it cleans more quickly and returns to its home base more reliably. Also, less cleaning required due to a different brush design.
Overall, I feel that I’m definitely getting my money’s worth out of the product, and it’s doing a great job at vacuuming. YMMV, but for me, it’s great.
Better to compare it to a cleaning service. Say $30 an hour, and the roomba does the rough equivalent of an hour a week when you let it run daily. 4 years (208 weeks) of $30 is over $6000, ten times the price of the 780. Even if the roomba does the equivalent of 6 minutes of maid-time each week, that's not bad.
But the cleaning lady doesn't take nearly as much time vacuuming as a roomba I just don't think it's a comparable service at any level or that it can replace a cleaning service - it can supplement it by vacuuming more often.
I like my Roomba, as the way I like my dishwasher.
Can I personally clean the foors better? Yes. Can I personally clean my dishes better? Yes. Dishwasher sometimes turns things upside down, doesn't clean properly etc.
But the main reason is that it saves me a lot of time, and it's good enough for me. Maybe not for you.
I've been thinking of what a more efficient and effective dishwasher would look like, and I think it would be something like: 1) you put dishes in it continuously and it cleans continuously, no start/stop cycle, such that when one is done eating, one puts the dirty dish directly into an opening that then slowly swallows it up (leaving dishes in the sink to dry out is what makes them so hard to clean). 2) a more effective filtering system that empties out automatically at a more frequent rate, so you can put very dirty dishes in and it can handle them without recycling dirty water and 3) ideally, robotic graspers that can hold each dish, conceptualize the shape, and actually scrub the dish.
Efficiency with power and water and a low level of noise are what I looked for when buying a dishwasher. It's hard to get these with the above requirements.
That has sooooo many more moving parts than a traditional dishwasher with one moving part powered by water pressure. And traditional dishwashers are unreliable enough as it is.
First of all, it's not the same company as roomba (irobot). Sounds like you had issues with your roomba (you don't say what they were), but I don't see how that is related to being "tricked."
My anecdata: we've had three since 2009, the earlier ones cleaned well but got dirt inside and eventually stopped working after 2 years. Our latest one bought 3 years ago seems better designed and more reliable, zero problems so far. Ours runs 2-3 times a week, mostly on wood and linoleum floors, and fills up with pet hair and dirt each time. We don't fiddle with the IR barriers, we just put a box in a doorway if we don't want it to roam, and we don't worry about docking, we just pick it up wherever it stops, empty it and dock it manually. Overall, it does a good job of getting most of the dirt in an area, but there is still the need to do a deeper clean once a month. Not affiliated, just a satisfied customer.
I think you are missing the point -- The Roomba barely works better than YOU cleaning floors. And this weeding robot does a crappy job compared to ME weeding my garden.
But our Roomba was a miracle worker when my wife was laid up in bed for 6 months with a shattered pelvis and I was also extremely sick. And while I know how, and have time, to weed my garden, this robot could help people who don't have the time or energy to keep up with such thing. And the stories don't even have to be that dramatic - in a family with 5 kids, there are better ways to spend time than vacuuming carpets.
Sometimes having devices like this isn't about the quality of the work - it is just about the work getting done at all, so your time can be spent on more important things.
I agree. Companies and consumers alike need to step back and ask themselves "How much time does it take to complete this task without this product? How much time on average will it take me to complete the task with this product?" Those are the hard numbers you should look at when buying a home automation product. I wouldn't be surprised at all if for Roomba, the numbers are 1 hour and 45 minutes, respectively.
The Roomba doesn't clean as good as I could, doesn't clean as fast as I could, and doesn't clean as cheaply as other products could. But I live in a house with laminate floors, two dogs, and a cat. Trying to keep up with their fur is a losing battle, so I'm more than happy to spend $300 once and have something that will run every day and keep the fur off the floor.
It's not replacing a normal vacuum, anyone who thinks so is fooling themselves. It's to keep your house cleaner in between normal deep cleans.
The term "weed" feels pretty outdated at this point. People consider many beneficial plants to be weeds just because they didn't put them there. For example, people remove clover, a wonderful nitrogen fixer, from their lawns in order to maintain their unsustainable monoculture of nonnative grass. Is a weed-killing robot solving the right problem? Modern agriculture works against nature rather than with it.
A weed is essentially any undesired plant that self propagates effectively. The fact that some of them are beneficial in some aspect is irrelevant if they are unwanted. The entire point of a garden is that it's full of stuff that some human wants.
Clover is certainly not what concerns typical gardeners. Is there a reason I should love the morning glories that are constantly trying to kill my tomatoes?
Clover is not necessary for healthy vegetables. Also he was talking about clover in the context of lawn, not garden anyway. So this robot wouldn't even be relevant.
The whole comment seems bizarre to me. It's like some strange gardening pretension.
It's not necessary, but clover is very beneficial for all sorts of plants because it fixes nitrogren. I can't reply to some other person but yes, it's good for vegetables, too, because more nitrogen means more growth.
Had some builders in who trashed the lawn and it needed covering quickly to reduce the mud. I threw down a clover and fenscue mix and within 7 days had a full lawn that is near bullet proof, winter and summer. I'd prefer not to have clover but it's extremely drought resistant and the bees love it.
Weeds as the westerner farmer understands them to be are often transplanted from other parts of the world and without a natural predator, and can take over fields pushing out the crop.
Mom fights a never ending battle with foreign weeds like knapweed (from eastern Europe) and Canadian thistle (from southeast Europe and Asia) that was introduced thru contaminated seeds. Every year she sprays and manually digs the weeds out of the ground.
Edit: also should point out that what we call weeds push out the local plantlife too. I also forgot kudzu, the weed that (or didn't? depends who you ask) ate the south.
As far as I know, the term is fine. A plant you don't want is a plant that you don't want, regardless of whether you appreciate why you should want it.
As shown in the video, plants you want to keep can be saved from the Turtle. 'Weed' is just a term people recognize as "something to remove from my lawn." If you want to save clover, that's a fixable problem. Which they fixed.
Gah. I have to wonder who thought the ieee.org website design was a good idea. 2/3rds of the page taken up with not the article, but a whole bunch of noisy, visually intrusive links to everything and its mother? The article sort of squeezed in on the left 1/3rd seeming almost like an afterthought.
You've both zoomed out in your browsers. Press CTRL+ a few times until you get back to 100%. Article takes up half the screen for me in Chrome+FF, 100%zoom @ 1440p.
No I haven't. Standard 1080p display, multiple machines, multiple browsers (Safari, Firefox & Chrome.) On all I get the same thing, the article is taking up barely a third of the page.
I'm not a huge fan of landscape fabric. It works for a season or two but over time it forms holes that weeds push through. You still have to manually weed but now you are less likely to pull all the roots up.
The previous owner of my house used a ton of it as a quick solution to a weed problem so they could more easily sell the house. I've spent a lot of time ripping out their quick fix. I had a nice surprise of finding a bed of sad irises under one.
> Currently I’m thinking that cameras connected to deep learning networks is the technology with the most promise to enable a new wave of low-cost robotic applications.
After his earlier comments about fancy sensors being too expensive I was surprised to read this. But I agree. Once some cheap neural net ASICs come out I expect a revolution in robotics will follow soon after.
I'm not sure if I understand your main thrust, but running data through a trained NN can be done in relatively few operations - it is training the network which takes a lot of compute resources.
Cameras are pretty cheap and already mass produced like crazy. The expensive part is the processing you need to extract high level information from the pixels. That's why we need cheap neural net ASICs.
This looks like it would be fine for a homeowner that wants a garden but doesn't like to weed. But looking at it from the commercial standpoint it doesn't make much financial sense.
You'd need to spend nearly $109,000 per acre and that is a pretty small vegetable commercial farm. In fact ten acres would probably be the minimum for one family to make a living and you'd spend over a $1 million!
You can hire a whole lots of high school kids every year for a whole lot less money. Doesn't mean it isn't cool technology, but then again so is a $450,000 Ford GT.
Still believe that commercial herbicides will soon be an endangered species. Didn't get many farmers to agree with me twenty years ago but maybe more would now.
I imagine once this is made to fit into the row spacing of commercial row-crops, and purchased at scale, larger operations would be able to deploy these at a cost competitive with herbicide applications. And once it gets even within an order of magnitude of traditional herbicides, the implications for organic crops could be huge!
Agree with you more than you do yourself. Herbicide costs are rising while technology costs are declining. First use will be organic crops but I think it will make sense for non-organic ones as well.
Once technology like this hits a critical mass, I'm not sure there will even be the need for a distinction between organic and traditional agriculture.
edit: Though I suppose there will still need to be distinctions made to make clear the differences between GMO varieties such as BT corn and the like. However, maybe by then the definition of organic will widen to allow for the incredible advancements that genetic engineering allows us.
There will be farmers who will want to perpetuate a distinction just as you have microbrew vs craft vs "commercial" brewers.
Once a majority of produce and fruit can be classified as organic from a chemicals PoV labor intensity may become a point of distinction in order to fetch a better price in the marketplace (I say that as someone who typically supports /buys/ organics).
I guess I don't understand the distinction you're talking about. Are you saying that people would be inclined to pay more for food they knew had more human hands involved in it?
It's a bizarre future, but I guess we're already seeing people pay 25x for a hand bag if they know it was made in a conflict-free village or something of that nature. I suppose its only natural that our abundance will begin to allow us to start making those sorts of luxury designations for our food as well.
It's exactly that distinction I'm making. Oftentimes automated machinery is better at making things than humans at making the same (furniture, cars, shoes, etc.) but people like buying "crafted" items for various reasons (vanity, support of craftspeople, local economy, what-have-you).
However, the main point is some farmers will find some kind of marketable distinction bigger operations won't be able to leverage (labor intensity or small production varietals "heirloom", etc.) Something which would allow them to fetch a better price and make them viable as a business)
I expect freshness and taste/heritage to be a greater selling point than pure labor intensity. A megafarm halfway across the country may use equal cultural practices, but the produce can't be as fresh as the farm just a bike ride away, and for many crops freshness translates to flavor.
Moreover, a megafarm would like to grow the latest and greatest in crop genetics, bred for standard beauty. Small farms are better able to grow the varieties that have been around decades, taste better, and are less perfect and ship poorly or are only exciting to a few people. Or, small farms can grow the newest small-breeder varieties bred for flavor or amazing exotic looks: I'm a big fan of Wild Boar Farms' new tomato introductions, but most people would never buy a mostly-black or an outrageously striped tomato.
You have a point with "heirloom" varieties which will not get the attention from large-scale farming.
The freshness point, I would contend that vertical urban/sub-urban farming could potentially deliver better with regard to some fresh produce and some fresh fruit/berries --not all, obviously, but many of the best selling items --better because they can grow more with more uniform product. The qualifier is potentially, if economics work out.
Obviously harvests which require large trees are out of the question --you can't commercially grow a mango grove or apple orchard in a vertical farm.
Not unless someone 'creates' another source of organic fertilizer because there won't be enough to go around.
You'd still have a need for insecticides although GMO's continue to reduce the need. But there's awesome potential for agchemical sales to go down and I say that as a former salesman of them.
Doing it at commercial scale is probably a much harder problem. If they're successful for hobbyist gardeners for a while I'd expect them to then move up to bigger operations.
The wheels look too tiny for the real world. Nice idea, though.
This guy invented the Roomba? He must have been really screwed by iRobot if he has to do a Kickstarter to fund his next project. He should either have enough money or the reputation to talk to VCs.
The "goal" on KickStarter is set as US$ 120,000 (at this moment it reached more than double that US$ 265,213).
So he asked to receive between 108,000 and 110,000 US$ (net of the Kickstarter fee of 5% and of the additional - absurd BTW - "payment fee between 3 and 5%).
This can be a rather large or a ridicoulosly small amount (of course it depends on points of view) but I propend for the "ridicolously small".
A (hopefully) succesfull robotics engineer in his fifties should have (or can procure through relations) that amount of money easily, and he is "only" a co-founder (and CTO) of the new company.
It seems to me - I have noticed it on other projects that ask for otherwise senselessly small amounts of financing - that Kickstarter is used only to give some visibility to the product/invention.
Seen this way the 10,000-12,000 US$ (financed by crowd BTW) are more "advertising expenses" than anything else, and even if at the end Kickstarter gets (thanks to the higher level of founding contributions) double of that, it is still peanuts for the amount of visibility it provides.
Yes exactly, and "having to do" a Kickstarter doesn't necessarily mean that he "must have been screwed", I think regardless of how much capital you may or may not have, there is no shame in raising funds, is there?
Might work on the East Coast or in water-intensive gardens elsewhere, but in California almost any reasonable garden will have a layer of rough mulch to reduce evaporation (and suppress weeds). Maybe this robot could navigate my mulch but I'm doubtful.
Joe Jones is one of my heros, I've still got my copy of "Robots: Inspiration to Implementation" that was such a great focal point for robotics in the 90's.
Tilling seems like something that would be pretty challenging though. The whole "weed/not weed" thing has a lot of interesting questions around it. I always thought something that would keep your driveway and sidewalk clear of snow would be the next bit opportunity to automate.
I can think of two very specific (for me) use case. Kill weeds in driveway pavers / patios where anything growing green is fair game and a Poison Ivy extreme prejudice terminator. I don't want to use pesticides since I have dogs and enough land to see the occasional rabbit, turtle, garden snake and I don't really want to harm them either.
pubmed is your friend: https://www.ncbi.nlm.nih.gov/pubmed/10854122/ is a good example. "Multiple lifetime feeding studies have failed to demonstrate any tumorigenic potential for glyphosate"
Roundup is about as safe is it comes - common vinegar is more toxic.
Well there's still a lot of back and forth on this from my understanding. Earlier this year some emails came to light from Monsanto regarding ghostwriting some research, aka they wrote an article and more or less had some scientists sign off on it as though they had done the research. They were also tipped off about the International Agency for Research on Cancer's determination for being a probable carcinogen so they were able to "prepare a public relations assault on the finding well in advance of its publication".
There is always the Monsanto conspiracy... That group is very vocal, but hasn't done a very good job of getting real research past peer review (getting their "research" into newspapers is easy and they do that all the time). Monsanto is evil (source: a personal friend employed by Pioneer - about as biased as they get), but getting past peer review is hard, even when you are evil.
Yes, but you can't ignore that the study you mentioned above was done funded by Monsanto, they paid Intertek to do it. [1] "Ten of the 16 scientists on the Intertek panel have been consultants for Monsanto in the past and two others are former Monsanto employees, according to a roster published on Monsanto's website." Paired with the unsealed internal emails from earlier this year, it's at least enough to raise an eyebrow at.
Unfortunately I do not think that peer review is incorruptible either. It is certainly important and speaks more for credibility, but especially in a case like this where there is so much money involved, personally I'm still waiting to see more information. I mean those emails just came out a few months ago, who knows what else will be found at this point?
The "if it's small, it's a weed" is probably ok for a v1.0 product. They might want to talk to Toshiba - they bought IBM's retail Point-of-Sale business a few years ago, and they had a product called "Veggie Vision" [1] that identified produce at the cash register via a camera.
Insert "Not Hotdog" joke here, but having a wifi connection to the cloud where heavy-duty weed classification software could be run would keep the device costs low.
>Insert "Not Hotdog" joke here, but having a wifi connection to the cloud where heavy-duty weed classification software could be run would keep the device costs low.
Actually I would like to insert here that time the connection was down for half an hour and the thingy eradicated all aunt Julia's plants, and it's not a joke (at least according to aunt Julia, who didn't laugh at all).
The article pointed out the inventor's book, Robot Programming : A Practical Guide to Behavior-Based Robotics. [1] Are there any other interesting/recommended texts for this domain?
The book refers to a website that seems to have been domain squatted by GoDaddy. An alternate .net tld is present, but unfortunately uses old Java applets. [2]
Tertill solves a big issue. Lots of ppl want to have a weed free garden and don't have time/ don't want to pull out weeds all the time. But I do hope, though, that Tertill is just a very basic starting point. I can immediately think of 20-25 improvements.
How long will weeds last if you keep decapitating them? I'd imagine they would deplete the soil even faster than just letting them grow(which brings it's own problems with shading out plants and propagation)?
It feels like everyone gets their own Martian rover to me. still a world of solar powered helpers for menial tasks is interesting. with low enough costs eventually you won't care about how many you need.
The main issue with just trimming is that sprouts coming up from roots of nearby trees have infinite energy and will get thicker every time you cut them.
Not to mention that the worst weeds also spread horizontally
Seems like insects and rabbits are a bigger problem for the typical backyard gardener than weeds. So if the robot has insect and rabbit shooing modes, maybe...
The fact that it isn't worth the effort to steal it? I mean, it's a nice demo, but it has to be better than that to be useful. They say it needs planks all around the work zone. That could be solved with a camera placed at some height. What if the soil has small ditches, as it is the practice when planting? It needs nice level soil to work, as well. It's not using computer vision, or not an advanced form of CV to distinguish good from bad plants. So it requires manual identification for each plant. In my opinion, it's not ready for prime time yet.
Yeah.. wow. Stop by a Home Depot sometime next spring and observe the mountains of bagged mulch they stock. Home gardeners and hobby farmers do indeed mulch. A lot. Where I am they stack mulch between gas pumps at every station. If that's too much trouble a single call will have a truckload on your property inside a week.
Simple weed killing robots of the sort depicted here will find their home on lawns. We're not all living in a coastal desert; ample fresh water affords nice lawns and that means squabbling with weeds.
I was just responding to the video, which seems to clearly be an unmulched home garden. So I am assuming from their marketing that is the application they are targeting.
You can use cardboard or a few layers of newsprint to broadly cover things, and then cover that with thick layer of straw or wood chips.
It has huge benefits for water retention, weed prevention, as well as providing carbon to be broken down by the soil life which will improve your soil over time.
I am planning on getting 75 yards of wood chips delivered to mulch an orchard we are putting in. For me the mulching robot will be a bobcat I rent to help spread it :)
For reference, 75 yards of wood chips (3 tractor trailer loads) + bobcat rental to spread costs $1000 in Bay Area. And provides years or weed prevention, water retention, and soil improvement.
Presumably you have seen those vacuum things that pump the mulch out?
They do all the motorway berms with them here. Take care you don't go too thick with the mulch as it burns very well and when fresh gets very hot. It's midwinter here and the inside of a pile was too hot to comfortably touch last week.
This is ridiculous but I'd rather see people going this route than spraying glyphosate or some other poison. That said, the root of this is a dangerous kind of hubris.
The idea of weeds is a tragic fallacy. There are some "weeds" that are selling for up to $15/lb (nettles, lambs quarter). Most weeds are beneficial for cooking and herbal medicine. And most any weed can have some kind of beneficial role in the garden.
Same thing goes for "pests". One plant's pest is another plant's companion.
Yet people are all but obsessed with exterminating misunderstood creatures from their local ecosystems. In my view, this robot is not so far off in the grand scheme of things from a robot that goes around killing other animals and even people, based on shallow, superficial misunderstandings of another being's potential.
I wish as much engineering effort went into understanding as it does into stuff like this.
Lets not get hyperbolic - folks pull plants for a local monoculture of beneficial botany. That's all. Its not a skynet extermination program or anything.
Heck, I pull volunteer flowers from my flower bed if they don't jibe with this years' layout. Its not about what's a 'weed'. Its about what I want in this here bed, and what I want in that one over there.
On one hand, I feel you. Actually, I think a "weeding robot", programmed well, is a much better alternative to the chemical/GMO program.
But weeding as an act of moderate human selection isn't what irritates me. It's peoples' blind acceptance of an entire false taxonomy of "good" vs "bad" plants/organisms. It's dualistic and dangerous in the long run. It's stupidity posed as common sense. It's a kind of eugenics that's directed not just at humans but at entire ecologies.
Right now what Monsanto is doing with chemical resistant plants is unconscionable. The glyphosate scheme is obscene. It's worse than some skynet extermination program. In the midst of a massive loss of the earth's sustaining biodiversity, we have unchecked corporate powers playing out obscene world domination plans. The stakes are too high. Why do people feel so nonchalant about the Roundup scheme? Why do we eat their corn? Cause they're only killing weeds.
There's a massive extinction happening right now. What are we doing? Programming robots to decrease biodiversity.
My parents used to make me weed our yard as a chore and a punishment. I still harbor an irrational contempt for the sow thistle you see all around southern California.
I heard someone, a successful ecologist or chef or landscape artist I believe, once say their parents wouldn't let them pick a weed unless they could identify it.
That struck me as a truly brilliant idea. I think it also helps combats that fallacy you've pointed out.
However, a concern that I'm sure they've thought a lot about is in it's technique. I really only care about making sure there are no weeds within, say, 8 inches of my plants (or put another way, my level of concern about a weed is proportional to it's distance from my vegetable or flower plant). This looks like it will weed everything except the 8 inches around the plant. Never mind things like cucumbers that spread out on the ground and have important bits that are short and can't be contained by the wire protectors. Still, an exciting place to start.