OpenAg Is A Personal Food Computer

When a device that calls itself a personal food computer lands in your timeline, what image springs to mind? A cloud-connected diet aid perhaps, advertised on TV infomercials by improbably fit-looking Californian ladies crediting all their health to a palm-sized unit that can be yours for only 199 dollars. Fortunately that proved not to be the case, and on further reading our timeline story was revealed to be about a computerized farming device.

The OpenAg Food Computer  from the MIT Media Lab Open Agriculture Initiative bills itself as:

“a controlled-environment agriculture technology platform that uses robotic systems to control and monitor climate, energy, and plant growth inside of a specialized growing chamber”

It takes the form of a tabletop enclosure in which so-called climate recipes to replicate different conditions for plant growth can be tested. It’s probably fair to say that in this most basic form it is more of an educational device than one for full-scale food production, though they are applying the same technologies at a much greater scale. Their so-called “Food servers” are banks of OpenAg environments in freight containers, which definitely could be used to provide viable quantities of produce.

The good news is that the project is open source, and their latest story is that they have released version 2.0(alpha) of the device. If you are interested, you can read the documentation, and find all the resources you need to build one on their GitHub repository. They page linked above has a video that’s very much of the slick PR variety rather than the nuts-and-bolts, so we’ve sought out their build video for you below the break instead.

Computerised agriculture has featured here more than once over the years. Just a couple of projects we’ve shown you are an autonomous farming robot from Australia, and Farmbot, a farming robot and Hackaday Prize entry.

[via Hacker News]

18 thoughts on “OpenAg Is A Personal Food Computer

    1. Pot, veggies, herbs. Dozens of soil and hydroponic setups have been documented here on HaD; one of the early uses of Arduinos was in fact for this stuff. Many of the projects have started with light control (the simplest setup) and added in control of temperature. Watering based on timers, soil sensors, or humidistats usually follows. The more advanced setups include some sort of fertilizer dosing system.

      This a-hole and his postdocs from MIT Media Lab are going around giving TED talks about it, acting like it’s their idea and like they’re forging new territory. It’s the usual MIT Media Lab snobbery / condescension – identify a movement, swoop in and act like it needs them to build a better, more professional version.

      “Even broccoli will have an IP address” was one of his talks. We couldn’t lampoon bigger d-bag that this guy. Look at the text from one of their videos:

      “openag is scaling up our open source platforms capable of controlling and actuating complex adaptive environments and digital climate recipes.”

      Translation: we bought bigger pumps and bigger lights and put them in a shipping container.”

      “flavor and the phenome”

      Translation: buzzwords!

      “virtual farm”

      Translation: buzzwords!

      “scratch climate creator”


      “cellular agriculture”

      Translation: ignoring the basic fact that agriculture works best AT SCALE and is primarily about harvesting solar energy.

    2. I just loaded up the BOM and did a SUM on the parts cost. Nearly $1,000 for about 2 square feet of growing space.

      ONE THOUSAND DOLLARS. That doesn’t include dozens of custom 3d printed parts.

  1. While it’s probably not efficient yet (cost to implement/payoff time here on Earth) this might be one kind of thing we want to send to Luna or Mars. Get them running before colonists arrive, and have a supply of food already waiting.

    You shouldn’t rely on shipments from Earth long-term. You want your primary food coming from something controlled and reliable, like these. Then, if you lose at the balancing game in Biosphere III, you at least don’t starve.

    1. Sending people to Antarctica is the best example for what will be required to keep people alive on Mars, and yes there is a small feel-good 200 square meter greenhouse in McMurdo. But the bulk of food at Antarctica is still shipped in with the large amounts of fuel required to keep people alive there.

      Antarctica (ref: )
      min: −93.2 °C [−135.8 °F]
      max: 17.5 °C [63.5 °F]
      mean −57 °C [−70.6 °F]

      Mars (ref: ):
      min −153 °C [120 K; −243 °F] at the poles
      max: 20 °C [293 K; 68 °F] at noon, at the equator
      mean −55 °C [218 K; −67 °F]

      Antarctica has a serious number of advantages when compared to growing food on Mars:
      1 light levels – the amount of sunlight reaching Mars is ~44% less than reaches the earth (inverse square law).
      (one solution would be to use lots and lots and lots and lots of 46% efficient solar panels to concentrate energy and power red and blue grow lamps, another might be using mirrors, or another would be to to grow more and let the plants grow 54% slower)
      2 enough atmosphere pressure (that water does not boil at ~10 °C)
      (many solutions, almost all involving energy to pump Martian atmosphere into pressurise containers)
      3 oxygen, plants do require oxygen at night time
      (many solutions, should be extracted from CO2 by the plants, but initially could be thermally extracted from iron oxide if required, it all depends on the scale of the operation)
      4 Nitrogen
      (can easily be concentrated from the extremely low pressure the Martian atmosphere ~1.9% N – Mars has ~0.6% of Earth’s pressure, so multi-stage pumps with some major safety features would be required.)
      5 water
      (not very many solutions)
      6 heat, plants require heat to transport nutrients.
      (see solution for light levels above)
      7 PGPR (plant growth-promoting rhizobacteria) is required for plants grow
      (brought from earth)
      8 carbon dioxide
      (could eventually be provided by humans, or initially pumped in and pressurised from extremely low pressure the Martian atmosphere ~96% CO2)

      There are two areas where Mars beats Antarctica the soil there on average is richer in potassium and phosphorus than earth, so that is a plus.

      1. Nice post.

        It reminds me that trees grow out of air, not out of the ground (seriously, they re made mostly of carbon, oxygen and hydrogen, and get that all from the air.

        I would love to rig tousads these up in progressively more mars-like circumstances, (low light, high CO2/low O2, low pressure)bombard a bunch of hemp seeds with radiation to “stimulate” evolution, and Darwin the suckers into something that can survive outside on mars.

        1. Personally, I’m kinda miffed that they didn’t hit Mars and Venus with extremophile bacteria and algae etc to get the ball rolling as soon as they could get a probe there 30 years ago.

          1. Humans have a poor history of invasive species. Kudzu comes to mind. I’d be worried that we’d end up dealing with a cascading imbalance from haphazard terraforming efforts.

            Makes me wonder if there is serious research being done for remote terraforming. My guess is we lack the bulk of the environmental data needed for worthwhile experiments but I have no knowledge to back that up.

          2. Problem with terraforming is that we destroy an aweful lot of environment, geology, and the natural history of the solar system that way.

            Plus the risk of xenocide

      2. I almost took a job at McMurdo for the NSF, but got married instead. It really is what I imagine an extraterrestrial base would look like, and also why the ‘ship it over’ philosophy that has ruled Antarctic and Mars planning is killing the opportunities to do more science in hostile environments. It will also bootstrap more opportunities with economies of scale in automation of personal or coop sized means of production meaning we are not left behind when the ultra-rich have compounding robot profits and the rest are asked to go starve NIMBY in a post-job economy.
        They had a very cool conex sized nuclear power plant at McMurdo, small enough to schlep it in whole on a C-130. It replaced literally tons of of fuel a month, but I think they hauled it away in the 70s. The energy density of nuke sure beats constant fuel tanker deliveries and less radioactivity & gaseous carbon into the environment than fossil fuels.
        The problem with bootstrappy stuff like this ag machine(too high intensity for temperate or even desert terrestrial economics?) is that we are close, like with our RepRaps but we do not have a closed production cycle form seeds in the ground for PLA and mining/recycling metal for other parts.

        1. As I understand it, the nuclear reactor was removed because they read higher radiation levels.
          After they removed it, they found out that background radiation levels were naturally higher there to begin with.

      3. “(can easily be concentrated from the extremely low pressure the Martian atmosphere ~1.9% N – Mars has ~0.6% of Earth’s pressure, so multi-stage pumps with some major safety features would be required.)”

        You have a really odd definition of “easily.” Where are you going to get all the power for that, the LEDs, the heaters, etc? How are you going to keep these very complex systems running reliably?

        1. You can look at problems as being really complex or you can break things down into lots of individually really simple pieces of technology.

          Compressing gasses is not very expensive or complex. There are expensive membranes that can be used to separate gases based on molecule sizes.

          Or fractional distillation, you freeze the gases and separate them using their different boiling points. All you need is mechanical energy ( ). Cryocoolers that use helium as the working fluid under extreme pressure are totally amazing devices.

          Or you do not separate the gasses at all and test on earth if compression alone with mixture of gases available on Mars is compatible with growing plants. So initially you provide the plants with (water, Mars equivalent soil, PGPR and) 96% carbon dioxide, 1.93% argon and 1.89% nitrogen at 0.1 to 10 atmospheres of pressure inside a sealed container and only provide light through red and blue LED panels, and then either cycled through a Mars daylight cycle or Earth’s. Experimentation on earth to reduce the final on site complexity is the key.

  2. Very interesting, and some nice build skills shown in the video, but essentially bullshit as no “real” food can be grown at that scale. People have no idea how huge an area is required to produce their daily nutrient intake, even if they don’t eat any form of animal products.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s