RepRap used for farming

Hello all,

I would like to ask here if there would be some developers interested in developing Farming bot which is in many aspects similar to RepRap projects. I am not saying it is absolutely the same thing, but people developing and contributing to RepRap forum could be a great power for inventing a farming bot used for home gardens growing food for themself and for small farms.

I am sorry if you will not like to see this topic direcly in RepRap comunity, I just want to try that, if somebody of you would be interested. I have been working with RepRap projects in the past and created couple of designs of 3D printers (XYZ and also delta printers). Some of them also added to wiki of RepRap comunity for others. I have been also trying to print from soil. Later on I start to investigate about other usage with similar technology and build couple of laser engravers and then also CNC machine. I am not technician expert, I just like DIY and creating things. So meanwhile I also found couple of project dealing with farming build on top of similar technology like RepRap is. Because I have a small garden and I like gardening and growing food for me and my family I can see some added value when some farming bot could help to grow a bit more food or keep the garden in good condition even we are not every day there. I described some of my ideas about such bot on following forum: here . The topic there is already closed but anyway if you would be interested you can write me here in the forum or send private message.

Once more sorry if such topic should not appear in RepRap forum. I would understand it. On the other side, if it would help to make community for such farming bot development that would be great.

I will be happy to see your comments here.

Thanks for your time and have a good printing day

Edited 1 time(s). Last edit at 12/11/2021 06:43AM by boylucky.

... a friend started with a similar project, where he used a RepRap-electronics to control a frame, moving a head with camera, a high power IR-laserdiode and some changeable "tools" over an area, inspecting the growing plant with the camera and "killing" weeds with the laser (by overheating the core) or nursing crops ... but cancelled for private reasons ...

I don't see any problems discussing/presenting alternative projects, based on or somehow connected to RepRap ... or you select one of the more specific subforums

Edited 1 time(s). Last edit at 12/11/2021 11:52AM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Hello Viktor (VDX),
thanks for your message. That project of your friend really sounds similar to this one. It is a pitty that he stopped it. Anyway thanks for this. There could be many approach regarding the tools the bot could use. Maybe it would be good to start in the beginning to just create a wheeled machine which would be able to orientate in the given space. Create it in form when it can autonomously move in the area. Then to start creating tools for it like seeding, transplanting of small plants, weeding, watering etc. I would like to use the bot in combination of no dig garden beds. That really looks pretty interesting this combination. Or even there could be invented specific style of plants growing specially for such bot. To make everything as simple as possible. But with amazing results which could produce enough food localy for all of us.

Thanks Viktor and have a good day....

Some quick thoughts having looked at some FarmBot info including [genesis.farm.bot] and a YouTube page at [www.youtube.com], and also because I spend a fair bit of time at our allotment*

First thoughts are that the FarmBot looks structurally flimsy, that it requires a firm base to be mounted on, that the use of extruded section is an unjustifiable expense when the required tolerances are somewhere between ±25mm and "dammit that is close enough", that the extruded section is of a form which could hardly be bettered for trapping soil and having bindweed growing through the channels, that the FarmBot itself will get in the way of doing necessary hands-on work in the soil, that the FarmBot will get in the way of having netting to protect the crops from the depredations of the feathered bandits, insects, and sundry neer-do-wells. That the FarmBot design is not well adapted to move from bed to bed and multiple FarmBots would be hard to justify.

Besides those few small problems, it is a very cool idea that has imagination going - not that I am able to offer more than ideas as I have too many projects on the go at the moment

To add a bit of structural integrity and get rid of the base I would recommend a structure such as the cages we use at the allotment, these are made from 50mm square treated timber and 20mm plastic conduit. The general plan is shown in the drawing below but these frames are 2400mm long, 1200mm wide and 1000 mm tall with wire bracing and plastic mesh to give added strength. They are light enough to be moved by one person and the smaller top dimensions (2000mm by 1000mm) allow them to be turned on their sides or even on their ends without falling over.



Taking this design as a starting point, the plastic conduit could be replaced with a steel conduit allowing an X carriage to be suspended on wheels that would run along the conduit. A Y carriage using a similar conduit and wheels would allow positioning above any plant.

Mike

* For those unfamiliar with these, allotments are micro-farms of about 1/8 of an acre where aging people spend their time trying to grow fruit and vegetables and successfully growing weeds and feeding opportunist birds and insects. Their main advantage is that it is a cheaper way to get exercise than a gymnasium without the sweatsuits and Lycra.

... another hint for this sort of "rough terrain" could be a combination of small camera-eqipped flying drones to first "scan" across the area an then, with some coordinates at hand, move the rolling/crawling "working bots" to the spots of special interest

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Hello VDX and Leadinglights,

thanks for your comments and ideas. You are absolutely right that Farmbot frame is not rigid enough and that the extrusions are making it more expensive as needed. Completely agree with you. I want to keep the project as simple as possible and also as cheap as possible but reliable for requited work. My thoughts was in a way to modify the Farmbot hardware a bit to make it more suitable for the wheels. Anyway my second idea, which I prefer, was to create new machine which would fit directly the no dig garden beds which are 75 cm wide so the machine with wheels whould be something about 1 meter width and 1 meter long, so let's say a cube. The top of the machine would be a platform for the moving tools and bottom would be used to drive the wheels. I will try to make a simple drawing of the idea as it is always better then many words trying to explain it. My idea was to get some experiences with the farmbot and use them in building a new machine. It will be a bit more difficult then just modifying existing solution of Farmbot, but I think in the end it would result in new designe anyway. If it will be used with no dig garden beds then I think it will be better solution.

Regarding the space orientation I found some interesting projects: here

Your idea regarding camere-equiped dron which would first scan the area sounds also good.

Both of these 2 ideas are amazing I would say. But my intention is to keep it as simple as possible to atract people to this project to build their own farming bot with as less as possible of experiences in the beginning. It would also require a months or years of precise work to enable the bot space orientation with just a 2 cameras on top of the bot which would automaticaly recognize very precisly the coordinates where it is located. So to avoid this huge development in the beginning and to lower a possible fail, I would go now a way where some infrastructure for the bot would be created on the ground. I have idea of plastic tubes used normally for water or electricity installations as something like a rail (or way keeping element) in the middle of 2 wheels on one side of the bot. There could be markes which would be red by the bot (barcodes or some other marks). From those markes the bot would be able to distinguish the coordinates in the area. The wheels would normally go on the ground but in the middle of the wheel there would be space for this plastic tube which would be connected to round to keep the bot direction. I will also try to do simple draw about it for better understanding. I think that this solution for the biginning will be more reliable to reach the goal of the farming bot, then directly trying to get the more sophisticated (but on the other side more comfortable) solution where the bot would learn the area by itselve from the cameras data.

I will also try to put some of the basic idea of the farming bot to a single document to express it a bit more for all the people whose would be interested in the development. I would split the development to couple of sections where people could help. I will also try to desctibe it in that document. But all is about discussion then as there can be much more reliable and better ideas then just my thoughts. The aim is to keep it simple and easy to build for everyone who would be interested. And also to keep it cheap. It should be also build from parts which would be mostly localy available (of course there will be always some parts which need to be get from outside but that should not be a problem as only standard parts also for electronics should be used. No special boards or parts. Just those which are widely available.

Thanks for your comments and enjoy your day.

Farmbot goes back some way, I thought it was a bit daft, especially when humans are cheaper(though maybe not for long), weather proofing & maintenance could be a pain. I think a smaller mobile rover would be better than some frame, perhaps one that pulls different trailers, you would think for the most time the, robot wouldnt have much to do after planting seeds(though it's a pity that task cant be left to nature), the robots main use would be soil monitoring & adjusting, bug zapping, the caterpillar killer... A project from years back called the Fergana Project in Uzbekistan had some crazy soil analyzing contraption, though it wasnt robotic...if only it was, it would have been ahead of the curve, there must be more up to date solutions these days?

The Farmbot that boylucky has pointed to is mostly for use at home and serves more to give fresh vegetables than any economic need. My interest is more in the allotments popular in the U.K. where I have some 18 vegetable beds of 2400mm X 1200mm as well as a bunch of fruit trees, grapevines etc. - about 1/4 acre or 0.1 Hectare in total.

Something like the Farmbot would be useful here where it could plant seeds in each bed before being moved on to the next one; in each case keeping a record of what had been planted. The most significant other feature for such a mobile robot would be weeding, but this would need a fairly advanced bit of AI.

A slightly different bot would be needed for watering as these would be needed on most or all of the beds. In some cases, this could be combined with LED lighting to extend the growing season and perhaps other functions.

Bots on allotments would need to be rugged and designed for long life and to be easily maintained. Problems encountered include frequent flooding which we get about every third year; High winds able to blow down greenhouses and polytunnels - these we have had two of in twenty years; damage caused by animals such as foxes, badgers, and squirrels occurs most years; damage caused by scroats of the "just having a larf" variety used to happen every winter in a previous allotment but have not happened in my present one.

So not as daft as MechaBits said, but something that needs a lot of development. Although this subject is a bit distant from 3D printing, the mind and skill set of the RepRap community is quite suited to this sort of endeavor.

I wish boylucky every success.

Mike

... for visual analysis and KI -- I'm actually developing a "moskito-sensor+shooter" with a stereo-pair of cheap NIR-cameras and a KI-board for some of the more "sophisticated" tasks - actually it's more meant for fastest possible detecting+targeting of moskitos, flying above the "shooter" range - here I have a 3D-position-calculataion every 0.02s and a "shooting" accuracy of roughly 90%

With more "KI"-analysis (what's much slower then) it could analyze the kamera images for known (or to "learn") contours/objects, obstacles, localisation or many other interesting "visual" information or acoustic input ... and "react" accordingly

The KI-boards are the Jetson Nano for roughly 100€ (or the faster Jetson Xavier NX for something below 400€) and the NIR-stereo-cameras from Arducam for maybe 120€ ...

Edited 1 time(s). Last edit at 12/17/2021 07:22AM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

I think a high powered IR laser for killing mosquitos could also tackle catapillars on the crops,egg laying moths, slugs, snails and even bloody badgers*

Mike

* O.K., I know that this is politically incorrect as badgers are supposed to be nice but if you saw the damage they have done to my lawns you would agree that zapping is too good for them.

... be carefull with high power lasers, controlled by software or KI !!

Our "moskito-shooter" emits not a laser-beam, but a "grid" of super-small droplets - around 0.03mm size per droplet ... the "field" is a raster up until 32x10 droplets, flying as fast as 5m/s (and I'm "tinkering"/developing with a novel shooting system, capable of speeds with more than 800m/s!!).

This fluid ist influencing/killing the moskito only by direct contact, but won't harm humans or pets - first for the extreme small amount of the active chemistry, second for the "basic intelligence" of the system, which will halt, if something bigger will entry the shooting range.

But -- if you really want to "arm" the bot with a laser -- my friend with his (cancelled) farm-bot got 3 IR-laserdiodes from me with 9Watts@975nm output-power each, what was enough to "cook" a plants core, when lighting it for some ten seconds, but not enough energy density (or heating time/accuracy) to kill flying beasts --- in the "laser-shooting-projects", I'm aware of, they used super expensive green (532nm, frequncy-doubled) NdYAG- or fiberlasers with galvo-scanners and some ten Kilowatts of pulse peak power to "evaporate" the wings of the target in some microseconds!

... could be, with my last high-power IR-diode-types with 270Watt@975nm this could be different ... but too, very, very dangerous to a "not intended target" along the beam

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

av u got a problem with moles Mike
When I say daft, I mean plants have been doing there thing for ages with just a little sun & water, but now we have to have robots to help them do it better, faster, cheaper?...and wipe out the income stream for a group of humans. I'm sure some ingenious mechanical contraptions could be made on the cheap to distribute seeds(If monsanto allows us to have some). Sure newt delivery, poly tunnel fixing needs something more advanced. Hydroponic systems are like growing crops in a lab type setup makes production line robotics easier to tag on.

Think of the security you would need to keep the robots & radishes safe, or the robots will be gone by morning. The Cabbage pickers will all have to re-skill as agro tech maintenance engineers.

Edited 1 time(s). Last edit at 12/18/2021 03:33PM by MechaBits.

... problems with "lost" (or stolen) robots was adressed some ten years ago with GPS- and some sort of network-communication (e.g. common telecom grids) - so you'll get at least the "last heart-beat", before the bot goes asleep ... sometimes even the moving path/location of the thieves

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
MechaBits
av u got a problem with moles Mike

When I say daft, I mean plants have been doing there thing for ages with just a little sun & water, but now we have to have robots to help them do it better, faster, cheaper?...and wipe out the income stream for a group of humans. ........................................

Moles I like, badgers not. Below is the sort of damage they cause.

I guess I can accept "daft" as not strictly necessary, I think I have become habituated to the taste of fruit and veg bred to taste good instead of the plants doing their own thing. On the employment of groups of humans, I can't think of a good invention that has reduced employment in the long run - they almost always create endless job opportunities that nobody would have understood in the pas. Take for e.g. the job description of "Influencer" which has become part of the internet.

Quote
VDX
... problems with "lost" (or stolen) robots was adressed some ten years ago with GPS- and some sort of network-communication (e.g. common telecom grids) - so you'll get at least the "last heart-beat", before the bot goes asleep ... sometimes even the moving path/location of the thieves smoking smiley

On thefts of robots, I haven't seen much theft at allotments, perhaps one spade was stolen about thirty years ago. Mindless damage by youths of the mindless kind is a bit more common so knowing where your wrecked robot is not helpful if the wreckage is just where it got kicked to.

Of course, the ultimate farmbot is an Asimo with a utility belt with trowels, forks, and dibbers, a Colt peacemaker in one hand for dealing with scroats and badgers, and a Wack-a-Mole hammer in the other.



Mike

... another idea, I'm thinking about testing, is a "cat-buster", loose based on my "moskito-shooter" -- the same stereo-IR-cameras and Jetson Nano (or Xavier NX) as with the moskitos, bit here a bit more "powerfull" water-canon, which won't hurt the cat, but should give some hefty reaction

For the badgers (or other unwanted intruders) this could be optimized by adding some "skunk-pheromones" to the water

Edited 1 time(s). Last edit at 12/19/2021 10:42AM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Hi all,
thanks for all your answers. I agree with you that some kind of smaller mobile robot would be much better then big frame Farmbot. I am definitely for this idea. I already started to make a testing prototype. About 610 x 610 mm from alu profiles. I already have prepared a moving part in X and Y axis (will need to add Z later). I will create the height of the robot for now about 500mm but will see later. Everything can be changed and will be depend on conditions and testing.

Regarding what MechaBits mentioned that the robot would not have much to do most of the time after planting the seeds. My intention is that the robot will manage also the watering. For now I have some experiences with water droping system from Netafim company which is great, but mostly can watering same kind of plants with same need of water. But it is perfect as it can be also used without high water presure from the pipe. Anyway because different wegetable need a bit different amount of water it would be good that the robot would be able to put just the required amount of water for each type of wegetable. It could make the robot a bit busy . But of course the aim is not to keep some machine eating some power, but to have healthy vegetable on the garden. I also agee with you when you say that after the seeding the nature know what to do and do not need robot which would try to help with it. I fully agree on that. But I can see a difference when I left a garden without my attention for couple of weeks. I could see that last year when I was not able to spend enough time regularly on one of our garden. On the other side too much effort also do not mean that there will be much more vegetable and food. So I think it is about to find a good balance. There are different techniques for growing like somebody mentioned here - hydroponic or even aeroponic etc. I think it could be easier to implement some automation and robotics to such type of grow. But I still believe that our planet give us the fertilized soil with all the microorganisms because that is the most efective and balanced think that can provide food for us and for animals. So I would like to be based on that.

Like I mentioned in the first note here. I would like to have the bot prepared for the no dig garden beds. More can be seen for example in this video here.

I think that the bot could be usefull for such type of growing.

From my point of view I would say that farming and production of fresh food is still vary labour intensive kind of work. I also think that the people there are often very purly paid. Then I would definitelly like many small local farms then one huge farm. I like the idea when we start to buy foods from our local farmers more then from producers thousands kilometers far. Or even better if we would be able to produce a bit more food by ourselve because then we become a bit more independent on the system itself. I would also like to attract more people to grow their own wegetable. People are almost to busy with their regular employment and they quit to grow their own food because no time for it. This could help a bit I think.

I do not want to fight here if such machine is even needed or not. I would be happy if people would have enough time to grow their own food because it can be very usefull in the future I think. Nobody knows what will come in the near future and it can be always good when people have a smale peace of land where they can grow their food by themself or with help of soma small machine. I think it could be a long discussion for separate thread or some live conversation just about if it would have some benefit to have such a robot or not. I still think it could be usefull and bring some benefit to us. I would also like to make this robot not dependent on the internet network. Much better would be when it can operate only with local resources and connections.

For now I think I would need to find some people localy to be able to connect our efforts to prepare at least some basic working model and later on to attract more pople to the development. I know I am definitelly not able to do all by myself. Anybody who would like to help is of course welcome.

Here is just a simple sketch of the current idea of the robot.

By the way, I am sorry for my crappy english...

Thanks for all your comments and ideas. And also have a good Christmas time and all the best to new year 2022.

... beside big flat areas -- you too can think about "vertical farming", where the plants are groving in vertically aligned shelves, where the water from the top rows is rinsing down ... then the "bot" could be a XZ(Y) system with a stable frame, only relying on belts or wire-driven linear gliders

Edited 1 time(s). Last edit at 12/22/2021 03:07PM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
VDX
... beside big flat areas -- you too can think about "vertical farming", where the plants are groving in vertically aligned shelves, where the water from the top rows is rinsing down ... then the "bot" could be a XZ(Y) system with a stable frame, only relying on belts or wire-driven linear gliders

There is at least one system using RepRapFirmware in this way.

Edited 1 time(s). Last edit at 01/28/2022 06:13AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Hi All, been quite a while.

I don't know about using reprap firmware, but this is the robotics section so I guess anything robotic goes?

I actually am a farmer, and some time back made a little venture into a project based on an old electric mobility scooter which fizzled out. I got it to the point of being a big glorified RC car, but it was useless off road and I didn't bother moving onto the RTK GPS stage. I can't find the the thread here anymore. Obviously I'm talking about a rover, not a fixed potted garden tender, so I'm probably off topic here.

As a farmer myself (beef cattle) I can tell you that if you want to save labor and help with the environment, robotic mowing and weed removal would be a good start. The bigger land stuff is being developed, but a smaller "swiss army knife" rover for various tasks on smaller acreages could be good.

If you are going to run something autonomously on a farm, you really need RTK accuracy. That's what they're all going with, and if it's your own place with stationary obstacles you might be good to go with that only and not have to worry about trying to make Lidar work.

Since I had my last dabble, consumer RTK kits seem to have gotten cheaper and better. Here's one that seems popular with the rover crowd:

[www.ardusimple.com]

Last time I tried to get a rear wheel drive with front steering rack to work, and I really don't recommend that. Skid steer with castor wheels at the other end seems to be what they're all doing. I really like the look of a zero turn mower as a base platform. All you need to do is remove the steering arms and use heavy duty servos on the linkages right on the hydrostatic motors. Much cleaner and easier adaptation. Here's a guy in the US that's got one working reasonable well:

[www.youtube.com]

I realize this may be a fair bit different than what you were talking about. I've come on here after hardly touching it for years and literally ran "farm robot" in the search, and this is what came up.

Hi Hobbymods,

thanks for your post. Interesting stuff... currently I have to put this project to something like on-hold as I realize that before I start with it I need to get done 2 more things. First one is to get a reliable budget and second is to get enough time to work on it. So I am trying to settle up a small business from which I could have some budget to support the development. I think that it could be really interesting for many people to have some garden helping tool which would help to grow them their own fresh food with just little of time. I agree with you that the weed removal could be a good start for the machine. To be honest I think that if such project should be really something that people will start to use, then there must be some growing technique made for it too. I mean I am still thinking that the no dig garden beds will be the right one with combination of the technical equipment. The bed with simple infrastructure. But of course there is so many other possibilities. Everyghing would be about the research... Thanks for you post and have a good day...

... think about the "verticsal farming" idea too! -- then you don't need big fields and the "robots" could too be much simpler/smaller

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]