Open Agriculture project - food computer


I have been interested in mushrooms since early my childhood
and I like to remember what it was like when I was a child in the forest looking for them, harvesting, cooking and preparing them for a nice meal.
Later I liked to take pictures because you could not collect them anymore. Chernobyl happend. The authorities feared that the radioactive failure had come to Central Germany and France. That’s why I was content with photography and literature.
It is very interesting what incredible powers mushrooms of different species have. They are biochemical all-rounders. They for example convert cellulose into proteins and other valuable ingredients. Additionally they extract toxins from substrates and chemically alter and detoxify them.
Apart from that, there are very spicy and expensive edible mushrooms. But some are difficult to cultivate.
When I saw the open agriculture project, I thought a little change in the design could create a suitable environment for hard-to-cultivate species to make it as easy as possible for everyone to use.
To give people an easy access to cultivate and harvest them.
Here is the original project
Open Agriculture Project

leaf detection example which could be modified to analyze efficient grow parameters

Hello I'm Chris and I'm doing my best to be a responsible proletarian

OpenAgriculture looks super cool.

My wife and I bought a small piece of land to grow native plant species in New Zealand, and we are discovering it can be a challenge to provide the right environment for their seeds to germinate. Something like OpenAg could make it easier to control and record exactly what is needed for each species… even the super difficult ones!

Have you seen farmbot? It is less focussed on monitoring and controlling the precise conditions, but it’s still an interesting project.


Yes i know farmbot.
Also super interesting, scalable and modular.
I imagine a future where harvesting and sowing is fun.
And a deeper understanding of biochemical and physical processes
becomes accessible to more and more people.

I am currently working on a script in C
to read a DHT11 sensor.

I can certainly also find something at Github but do not want to rush anything and first understand individual Input and Output scripts in detail.
to get a better understanding of open source sources.
which one can then tailor if necessary.

is that the right way to go?


I think it’s always good to learn how things work at the lower levels if you have the time!

While I implemented some simple neural networks from scratch at university, I still want to one day spend some time to try and implement newer techniques directly instead of just relying on the many neural-network and graph-computation frameworks available.


I assume you are developing in arduino(since most other IDE’s are C++), let me recommend an esp8266(arduino IDE compatiable) - that way you can easily wirelessly upload the data recorded to an API or to your own server.
And as far as I know DHT11 is digital, so no ADC is needed.


That is a very good idea.

Thank you. I will look if I can hook up the other sensors and activators as well on an esp8266.

That could be a much better way to build the Boxes in a much more modular way with less wiring.

I think this is an almost similar hardware .
which I have to modify to C,C++ for that special purpose and build up on it.
Some parts have to be added.

  • a cooling mechanism (maybe a water cooling system for Pc’s)
  • a mechanism to almost sterilize the box before and after it contains a specific Substrate.
    before the starting inoculation.(Ultraviolet light UV-C)

+ this brain written in python to set up the environment controls.

It seems that you are familiar with electronic parts. If you have some specific idea’s …thanks for sharing.


It seems like a really cool project, let me know over private messages if you are stuck somewhere with the hardware.
If you are talking about parts for the Fermentabot that are 3v3 friendly for the esp, then I would suggest the following:

air temperature = DHT11
humidity = DHT11
substrate temperature = Waterproof DS18B20 or any probe

I would skip these sensors - these are pricy compared to everthing else.
oxygen = LFO2A4
carbon dioxide = MG811 CO2 Carbon Dioxide
The sensors above is 45~60 USD a piece

pH = Liquid PH Value Detection BNC Electrode Probe
Electrical conductivity = two carbon rods in the water, i assume to check mineral build-up?

General advice: for power use switchers not regulators (energy waste), logic level converter (if 5v to 3.3v or vice versa). get some adc chips or adc breakout boards if you can’t etch PCBs) I would probably rewrite the program if I were you, takes less time in my experience. Good luck - share some photos of your progress.

Remember most gas sensors have to be changed every 3 or 4 years or so, as they corrode/degrade, especially in a humid environment.

Use for parts, they are cheap and you wont regret buying there :slight_smile:


Thanks for your suggestions. I really got stuck and it took me a while to get an overview of how and what. I’ll continue with that project more serious now and get me the components that are needed the next few weeks. And meanwhile write a prototype project map. I’ll post it here then. I think most probably it is better for me to have some development pressure to fulfill further tasks :smiley:

Why it takes so long…


Agriculture projects from a different perspective. We are commercially launching our AI platform that takes images from drones on large commercial row crop farms in Canada, Brazil, Paraguay and Eastern Europe.
Training AI to indentify Disease, Insects, Weeds and deliver prescriptions according to thresh hold provided.
any similar work being done here or collaborations?