Nature is beautiful. That's only true when looking at it from the safety of a window or briefly exposed to flesh in relatively clean areas. When you get up close, especially to grow things, there are all manner of critters that can wreak havoc on your immune system. Especially since we're far more sheltered now than we used to be. It's one thing to want to be demolished by weather, but quite another to experience starvation or food poisoning.
I'd like to grow a few things in addition to some basic gardening, but I'd also like to try my hand at introducing a level of automation into it. I don't think I'll have an actual "farm" in the sense of producting enough to sell, but I like the idea of supplementing what I need with a little extra as a buffer. I'll be extra careful not to lose perspective along the way.
This video is a few years old, but it is a great example of something I'd like to try eventually.
That's a beautiful illustration of what's possible with experimentation and drive. I wish Eryk Maundu and Kijani Grows all their well deserved success.
Hobbyist tech has improved quite a bit since that video was made and the Kijani Grows approach to aquaponics might be the way to try things small-scale. Especially for someone like me who lives in an apartment. Until I can actually secure land to try more expansive things, I'd like to experiment as much as I can in miniature.
While I like the idea of getting a pre-made "kit" to start gardening, I like starting from scratch even more. Plus, it's a great opportunity to learn what I can do with most of these components and experience more about the approaches I can take versus what actually works in the real world.
In that original project, they were using an Arduino Uno as the core with several relay boards and communication shields. This approach works for the most part, but also introduces several points of failure. In later revisions, it seems they have already replaced a lot of the disparate systems with a more integrated controller board that runs OpenWrt. That's a very sensible move.
I think a balance of passive tech and active electronics might be the way to go.
Water levels are probably best left up to passive balancing as much as possible, while moisture and temperature monitoring, fan control(s), and notifications should probably be automated. In that regard, I think there can be two main components; A core device which handles status reports and maybe notifications over the Internet and lower power devices that handle triggering relays and such. This also decopules the smaller relay controlling devices from the Big Bad Internet. Safety first!
A similar approach to that phone booth idea might work here as well. Leaving notifications to a central, hardened, device while the rest keep running their own tasks adds a degree of decentralization and resilience. This decoupling also makes it easier to power each gardening "unit" independently via solar panels and batteries (relays and controllers don't need a lot of power, but pumps and fans do). So if a single board and its independent system goes down, it won't take the whole farm with it. And replacing a single controller board should be less expensive than a bigger device.
A farming "unit" might just be a single greenhouse the same size as my cabin design. If I can build several on one plot of land, I can also diversify what I grow. I don't know if my first attempt should be aquaponics, but it's an option. It will obviously be much more involved than a typical herb garden.
Arudino boards are more capable now than the models available 8 years ago, but another option might be the ESP32; The same one I was originally planning to use for my DIY word processor. It may be a tad too underpowered for that task, but handling relays should be far more simple and straightforward.
And it comes with built-in WiFi (with WPA2), which gets rid of yet another another physical connection to a control source and obviates the additional shield board if I went with an Arduino.
As in the Kijani Grows project, the control and notification source should probably be something that runs OpenWrt. Router software should be easier to harden if I actually do decide to let it send me notifications over the Internet. This way, I can independently update the notification system from all the myriad controllers too. A simple router that already runs OpenWrt I was looking at was the GL-MT300N.
What really caught my attention in the new V2 are the two IPEX connectors for external antennas and the power requirements. The original package doesn't come with external antennas at all, but there's no reason to keep it in the original package if I plan to give it some margin of weather resistance anyway. It might be interesting to see if I can put all the ESP32 controllers on a dedicated WiFi LAN. The router board itself can connect upstream via ethernet and run on solar too if needed since it needs 5V at 1 amp and only uses 2.75 watts at most.
I'll need to check out what Aruduino IDE projects are available for the ESP32 before I can truly explore what's possible.
The entire concept of aquaponics is fascinating to me since it mimics a food cycle at least part way. There are extra bits and bobs for sure, but it also makes food variety possible, which I think is overlooked too often when it comes to distributed, sustainable tech.