Growing your own food is a fun hobby and generally as rewarding as people say it is. However, it does have its quirks and it definitely equires quite the time input. That’s why it was so satisfying to watch Farmbot push a weed underground. Take that!
Farmbot is a project that has been going on for a few years now, it was a semifinalist in the Hackaday Prize 2014, and that development time shows in the project documented on their website. The robot can plant, water, analyze, and weed a garden filled with arbitrarily chosen plant life. It’s low power and low maintenance. On top of that, every single bit is documented on their website. It’s really well done and thorough. They are gearing up to sell kits, but if you want it now; just do it yourself.
The bot itself is exactly what you’d expect if you were to pick out the cheapest most accessible way to build a robot: aluminum extrusions, plate metal, and 3D printer parts make up the frame. The brain is a Raspberry Pi hooked to its regular companion, an Arduino. On top of all this is a fairly comprehensive software stack.
The user can lay out the garden graphically. They can get as macro or micro as they’d like about the routines the robot uses. The robot will happily come to life in intervals and manage a garden. They hope that by selling kits they’ll interest a whole slew of hackers who can contribute back to the problem of small scale robotic farming.
Full scale agriculture is turning to automation to reduce labor costs in an ever increasingly competitive market place.
Delivery methods are more likely to be drones than some form of CNC because of the distances involved.
Targeted water delivery is accomplished by moving overhead systems something like this.
Of the manual processes, seeding is the one that takes the least time. It is only done once per crop and is a lot easier than harvest.
A version two with wheels and an ability to navigate has some chance of scaling up to the real thing.
They have the tool carousel idea down pat. Some suggested tools – ground penetrating moisture and alkalinity probe, Camera, Insect trapping device for later examination, temperature, UV measurement. etc.
“drones”? Ha! More likely they would modify existing sprinkler technology than that half-baked idea.
Obviously drones would be completely ineffective for water delivery.
Drones are used where insects are detected and deliver boxed predictors to combat the unwanted insects so it’s not a half baked idea. It something that has been trial and found to be effective and has been adopted in many places.
It’s not surprising that the use of drones has expanded as they were already in use as a monitoring tool.
What a nice, thoughtful comment and reply, Rob. I appreciate your contribution far above the flippant remarks of Gravis! Thanks for sharing.
drones are currently used for crop inspection, checking visual + ir to determine lots of factors, like photosynthesis for example. There was already an article describing this http://hackaday.com/2016/05/09/drones-what-exactly-do-they-offer-agriculture/
Farmers have been moving toward “Precision Agriculture” for a couple years now, but the FAA ban on large commercial drones has really put the halt to development. As soon as those restrictions are lifted you can expect to see drones with a leaping deer on the side!
I have an idea that is basically hexapods with a sprinkler and water turret on top of a water tank. The swarm of bots could water the lawn or garden and return to an automated spigot to be refilled with water.
Way beyond my current building ability though.
I’m liking this idea, but what if instead of a sprinkler we took one from the mosquito and inject a “proboscis” (a hollow metal tube with a point), plunge it near the root ball (this would require plant knowledge, learning and maybe a strain guage) and water them individually.
Of course that would open happen if the moisture sensor that was pushed into the soil beforehand said it needs water. You could then analyze that for water usage and good places to grow (maybe it isn’t as dry in one area compared to the others), etc.
Broadcast water delivery (sprinkler or drone-drop) is horrendously wasteful when compared to directed methods like drip lines. Even better at water efficiency are the various ‘ponic methods but tend to only be economical for specialty crops (salad greens, herbs, &c).
Add in that powdery mildew is a huge pest to curcurbits w/ damp leaves and broadcast watering looks even less desirable.
Since this is aimed at backyard and urban growers even drip lines or sub-irrigated raised beds (http://www.growtolearn.org/wp-content/uploads/How-to-Build-A-Sub-Irrigated-Raised-Bed.pdf) look like a better solution to the water problem.
As mechanization becomes required/integral to your farming these solutions become less manageable or outright impossible. Pulling 40 acres of drip lines before the combine comes through just isn’t feasible and a sub-irrigation would get compacted by the same equipment.
Drones really shine in being able to cover a lot of ground quickly. Using them for soil surveys, crop survey ( ie; is this patch getting enough water, is this sector suffering from some blight) and perhaps even targeted crop dusting. To date their payload capacity is limited compared to other means of flight so using them as glorified watering cans seems non-ideal compared to simpler existing means.
I’m really interested in seeing how they manage weed detection, maybe it’s on the .io page but I didn’t see it. A fun exercise to be sure, but weed matting or dense polyculture/no-till gardening is a much simpler solution to the problem. Light colored weed matting has the added benefit of reflecting light and increasing yield for many fruiting crops.
@Garrit
Paragraph 1, line 2 “…it definitely equires [sic] quite the time input….”
I believe this should be “requires”
Great idea but end effector needs to be tied to ropes and have three towers to move it around delta-ish style, would massively reduce the materials cost per area. That and it doesn’t really target the major issue of soil farming which is the soil itself. With poor soil you will get poor harvests, should be plucking the weeds and dropping them in the compost bin.
I like that you could use organic teas with the fertilizer system so big plus there.
Their weed detection is what I want the most and hopefully it’s ready by the time I can afford some land.
That would work, you dont need sub mm precision and it would allow for a massive scale up.
The soil problem could be monitored with camera’s I understood that soil quality isnt uniform even in a smallish field.
by monitoring the crop you could drop fertilizer/ pesticides where needed.
Four towers, not three. Think area coverage, a square is better than a circle.
what is the square footage of the farm the robots can manage so far?
1.5 x 3 meters is what they are shipping, so almost 48 and a half square feet.
I’m thinking this could be easily adapted to conduct vertical farming. In fact, it would be perfect for it. Each section can be pulled out of the stack to be worked on. It gets pushed back. The robot elevates upone. Pulls the next section out to work on. This help to keep it from being idle, it could work a larger amount of crop, it could also allow for less energy use per plant, and use less water if the stack allow for some of it to drain down.
If only….
This is brilliant. I am curious as to longevity/MTBF for this design. Its tough being outside.
i wonder what opinions someone who is actually currently into gardening has. because as a person whos done a fair share of gardening 20 zears ago as a child, this seems ludicrously overhyped….
You bring up a good point which is the general scourge of automation and robotics. I’ve had a small veg and fruit garden most of my years with cut flowers and containerized systems popping up here and there. In all I would say I have about 35 seasons in the garden.
I’ve been working with robotics for about ten years and I see many places where automation can fit into commercial and consumer food and material production. It doesn’t take much critical thinking to see why many of the ideas presented in the industry are very, very tedious and difficult problems.
In an era where critical discourse online consists of experts being pestered by pipe dreamers who do their “research” and return a few Google results it is no wonder many developers hide from mainstream forums and only reach out personally or on well-moderated professional forums.
I have not read all the documentation about the project so I don’t feel like I can give an informed opinion on the whole. The fact that this is an open source hardware project gives me pause since that is rarely done well unless a large organization opens the design of a mature product so it can be produced cheaply or used outside the box.
I have been burned by companies who claim to be non-profit and open source only to have them refuse to provide essential documentation to me which is clearly available under the license. It was not a surprise when I found out the claim of non-profit status also proved to be fraudulent.
From a technical standpoint I see many problems. Using watering as an example I can think of a number of problems probably not apparent to some of the fans of these ideas. The sheer volume of water is underestimated and the cycling of drone irrigation from a single source point is hilarious. Soil taxonomy is also a concern as is runoff; all soil does not water the same. Broadcast irrigation for many crops isn’t an an option due to disease like powdery mildew mentioned by another reader, a gardener or farmer would know this.
It’s asking a lot for people to actually have some hands-on experience before spreading what accounts to uninformed speculation on the viability of a certain technology. For many small developers it’s a terrible psychological burden and could lead to delays in bringing the tech to the marketplace. This is true of readers, commenters, writers, and publishers; often self-righteousness and stubborn refusal to be wrong can have the potential to kill a project. I am speaking from hands-on experience with that sort of added social burden.
It comes down to the old saying; “In the mind of a beginner there are many possibilities, in the mind of an expert there are only a few”.
the more i think about it, the more i seem to come to the conclusion that most effective would be growing pods for individual plants.
my reasons are as follows:
water runoff part a:
plants dont absorb everything you give them. they absorb certain minerals more than others, based on temperature and momentary physiological needs, as well as the particular ratio of minerals to other minerals currently present aroud the roots. this means that those who do not get absorbed accumulate, and interfere in various ways: ph changes, solubility of needed minerals changes, etc…. its a cocktail problem – the mix starts behaving in different ways than the individual components, often in unpredictable manner to the layman gardenere like me. so whatever is not absorbed needs to be evacuated.
water runoff part b:
plants also secrete things into the soil. stuff that prevents nearby plants of establishing a foothold, and also encourages other plants roots to develop. sometimes both phenomenons happen on other plants from their own species, sometimes from other species. search google for studies on rood exudate.
water runoff conclusion:
you cant mix and match any plants in a single pot, and you most deffinitely cant water some plants with the water runoff from other plants. careful consideration is needed depending on individual setups.
water runoff addendum:
sometimes you can use biological filters that are essentially high surface area materials to house bacteria that decompose certain root exudate chemicals, to treat water runoff and reuse the water. sometimes you need aerobic bacteria, sometimes you need anaerobic bacteria, sometimes you need both. in either case, special precautions are needed to make sure these bacteria dont escape their reaction vessel containers…
which brings me to the second major point:
disease spreading and containment
in an age where a lot of people want to use as little pesticides as possible, and when a lot of pesticides are rendered inefficient by means of tolerance development, isolation is probably the key.
having a mold establish a colony in the single container you use for watering all the plants, or any other disease, really: spores on the end effector that is used to apply various treatments and mechanical processes to all plants is the shortest way to spread any disease.
larvae, spores, eggs are now free to move around with speeds far greater than the nature has bestowed them with, by hitching a ride in the centralised technological system you use to maintain your garden with….
ah, the HORROR!
alas, having isolated growth chambers is a huge initial investment, and very hard to diy. dedicated systems used in plant research are ridiculously expensive, for the privilege of having air particulate filters that can filter out viruses and spores and such…. and have consumables that are equally expensive to maintain for the hobbyist.
i am VERY VERY interested in what kind of breakthroughs come from home gardeners and technology hobbyists, but from that to a food revolution… damn…. its a long way to go….
to detail a bit on the cost: just to monitor humidity for a single plant you need a couple of bucks per plant. initial investment. doesnt sound like much… but come to think that for a single persons consumption you would need at least a couple dozen for temperature alone….
and lets not forget that the most basic measure of nutrient concentration is conductivity, which you would have to monitor both at the entry point and at the water runoff point with some precision megohm-meters that are sufficiently noise free… so add a couple bucks for that too, times two (entry and exit), times the number of plants.
and then… then we start looking at the ph, which at a decent hobby level, if you take the high gain amplifier, and the probe, is, even in volume, easily a couple tens of bucks per probe system, and then you will have to change probes periodically every so often because the pores in the glass last only so much…. and youd have to store the probe in buffer solution between measurements, and rinse it off cleanly with distilled water before and after measurements…. and…..
i already got a headache. this gets expensive FAST!
It’s a marvelous project ! As a tech and nature enthusiast I am really impressed the amazing job being done on this project: thanks for pointing it ! I particularly appreciate the idea for watering plants precisely, and this could be a good alternative to the dripping system we own (every year, we have to modify and adapt it to the plants we want to grow).
At home, I grow some vegetables (potatoes, pumpkins, tomatoes, …), some fruits (grapes, raspberries, strawberries, oranges, lemon, peaches) and others (hops for my bier, liquorice, passifloras … ). We get a lot of pleasure growing and taking care of them when coming back from work. My children love to harvest them with me and appreciate their taste compared to industrial vegetable we can get at the store.
I have an automatic water delivery system so everything grows well (using rain we store in huge tanks).
The biggest time consuming task that remains is to remove weeds: as the soil is very fertile, they grow a lot and slow down the “good things”. I suppose it would be a real challenge to detect weeks using, for example a camera, but it would be definitively be the best function on such a device.
I think that a good improvement to this system would be to have automated pest control. When I grew exotic plant indoors (legal ones), I used so mist system connected to an espresso pressure pump to deliver organic pesticides (mainly menthol, camphor or coffee) in case of bug invasions (mainly aphids and red spiders). When I detected such bugs, I programmed my system to automatically deliver pesticides each day at dusk (some minutes after the lights were off) for a given period: this gave better result than by hand (using less pesticides), because it was more regular (the computer does not forget to deliver the pesticides, and deliver the right amount).
It seems like a nice developing in the path to some automatic food production, in some place without enough manpower. Like, a space colony, or even something buried under the Arctic.
In other normal places, it may have some uses, but general adoption could be more of a problem than a solution.
Yeah, I’m liking to see the tech develop, with a view to those purposes, but spending several hundred bucks to raise $20 worth of veggies a handful of times doesn’t seem very viable to me at the moment.
Ppl talking about the weeding has given me “ideas” though.
This is a truly terrible idea.
It reminds me of the guy who “came up” with the idea of using peltier elements to make electricity from campfires to charge cellphones. Of course, by the time everyone realized how bad of an idea it was, he had already raised over a million in funding.
Gardening an area of that size is really of no consequence. It would take much more effort to build this ridiculous thing.
If you want to just magically have vegetables without putting in the time or work, then go to the supermarket.
It saddens me to see negative comments on here, but this one really deserves it.
After thinking about it for a while longer, I regret what I said.
I still think this is an inefficient method of growing crops, and I think it doesn’t make sense, but none of the projects I work on make any sense. It was hypocritical of me. We all work on the things we love and care about, and they don’t need to make sense to anybody but ourselves. Sorry
It makes sense if you consider that you might have no time to weed or water plants for days or weeks (e.g. you want to travel, or your work demands so). However, irrigation is simpler with capillary systems ( like lines, etc) rather than drones or other solutions.
Cover crops, weed mats, and poly culture are all better (cheaper ?) ways to avoid having to weed every day.
It would be excellent for vertical farming on the side of buildings.
Farmbot is going to change the world.