In glasshouses a reasonably wide variety of crops can be grown hydroponically including tomatoes, cucumbers, eggplant, lettuce, cannabis, peppers. This Nat Geo article [1] has some interesting data/analysis on greenhouse growing in the Netherlands. The Netherlands are the leaders in high-tech greenhouse growing. Approx 0.2-0.33 of the world's high-tech greenhouses are located there. The biggest costs are financing/depreciation, labour and energy. The balance of these three costs changes significantly depending on your geographical location.

I am the founder of a startup [2] based in the Netherlands and backed by Founder's Fund focussed on scaling high-tech greenhouse growing as a method of crop production. I have both a technical background in Reinforcement Learning and have spent time training as a tomato grower. Happy to answer any questions anyone may have at [email protected].

[1] https://www.nationalgeographic.com/magazine/2017/09/holland-...

[2] http://optimal.ag

You can probably grow just about anything, but things with a long growing cycle, big root systems, heavy plants prone to falling over if not well-anchored, etc., are more work. So, fruit trees are a challenge, for example. They end up needing pretty involved scaffolding; I'd imagine that would be a particular challenge to implement with a robot.

Right off the bat they compare the production output of their robot interior farm with a comparable exterior farm "5 times the size" - which is another, more subtle, form of BS - how does it compare with a standard "meat powered" indoor farm? Production from hydro/hybrid is higher than traditional outside soil only.

>> "What we found is that the two biggest costs in the indoor farm are labor and electricity"

this may be true but it doesn't mean the costs of outdoor farming are lower, plus the added cost of land (good farm land is expensive to own or rent)

"plus the added cost of land (good farm land is expensive to own or rent)"

Urban land is much more expensive per unit. not only that, but you have the Capital expense of the machinery and buildings, plus the operational expense of heating, lighting, ventilation, and fungus control.

http://blog.zipgrow.com/indoor-hydroponic-farming-costs-prof... has a good overview of the costs. $110k for 500 square feet. excluding building. Then you have the opex for lighting which is $1100 a month. You'll need heating in winter, which will similar cost.

To put that in perspective, https://www.uklandandfarms.co.uk/rural-property-for-sale/eas... grade II soil. $80k, buys you 211266 square feet, with change left over for tractor, plough, seed and tillage hire/buy.

Moreover, its not an asset that will depreciate, unless one lets the soil blow away/don't improve it.

Airponics is even better, but requires high pressure pumps. You can grow pretty much anything, as long as you feed the right nutrients and control the temperature and the pH.

"Aeroponics" is indeed a good option but it does not need the "high" pressure pumps. We can pump with very low powered pumps. We, actually, formulate our own nutrients. We are about to start testing our automatic IoT solution to mix nutrients, measure the temperature pH and EC. Right now, we do the testing manually and we can slide the EC by a factor of 0.1 easily.

For those interested here is a video of our lab. Our structures are vertical for our research. However, our commercial (big) deployments will be flat or an "A" system. Right now, we are leveraging the abundance of sunlight in the tropics (India). We can produce 10x the traditional farming without any stacking but at a fraction of the cost of other hydroponics/Aeroponics setup.

Photos https://photos.app.goo.gl/XytrQYsQ0UaSKozV2 Locavore Demo Tour https://www.youtube.com/watch?v=U2CRI61_DlQ

Awesome work! I'm experimenting with a custom built, mostly-computer-controlled aeroponic myself, but it is way more on the hobby scale than this work you are doing.

One question - do you have any kind of automatic way to clear clogs in the sprayer heads?

As replied earlier, we tested and realized we can do much better by dripping it from top instead of the mist. We will also be doing flood-n-flush kinda method in the much bigger setup. Cleaning was one thing we learned the hard way and so the reason for our modified method.

We will still be using mist spray method with other crops that has less dense rooting, etc.

You could... detect pressure build up -> assume clogged

set pressure 0

open a solenoid connected to h2o2 with enough elevation to reach the clog

detect flow increase ->assume unclogged

close solenoid, set pressure 50 psi (or whatever)

This might best be done with a hydraulic circuit rather than electronic sensors, by the way. Ie. to detect pressure build up have a secondary line with slight resistance before the nozzle that when pressurized triggers other hydraulic elements which cut pressure, etc.

Woah, very neat setup. That looks more like a vertical NFT setup than aeroponics though. Is there a fine nutrient mist that I can't see? or is it simply pumped to the top then drips down the edge?

Yes, the structures are vertical NFTs. The method is borrowed from aeroponics. In our lab test, we wanted to source things locally, reduce CAPEX/OPEX to the lowest we can and have the best ROI possible. We realized that instead of spending additional motor horsepower and thus more electricity, we can reduce it by pumping up once and then dripping them from the top. The mist/spray method also clogs quite often and is more cumbersome to clean. Our next step with both Flats and "A" systems will use something similar -- perhaps a "flood and drip". We have also realized the benefit of the "Dutch Bucket" system for growing more common food/plants.

I looked into it, but it seems pretty risky? Especially since if u miss the schedule to most every hr the plant dies very quickly