Ask HN: What's happening in agricultural technology?

194 points by tabeth ↗ HN
Not sure if there are any experts here, but if there are, I'm quite curious about the following:

1. Do we have any technology (independent of cost) existing today to have a fully automated farm? Meaning, set everything up and it'll just spit out yield? This would include taking into account weather, planting, removing weeds, etc. The input into this "system" of course would be the "ingredients" such as soil, soil, and of course, sunlight. [1]

2. What's the maximum yield we can get out of an acre these days? If I wanted to buy, say, an acre (for food) and start a small town in southern United States and feed 1000 people easily, is that possible? For simplicity let's say everyone is vegan.

[1] Here's a snippet of research I've done that may be of interest to readers.

Air and soil: There's been some advancements in sensors which is an obvious requisite towards the auto-farm. However I haven't read anything recently about this. Most recent was (http://gizmodo.com/swarms-of-soil-sensors-may-help-farmers-water-smarter-d-1713098054)

Robotics: I know they (farm bots) exist (http://modernfarmer.com/2013/08/5-robots-on-the-farm/), but I haven't heard of them actually being more productive than existing high-yield solutions. I do think in the end they'll be superior.

The single most advanced modern farm I've heard of is: http://seedstock.com/2013/01/24/indoor-grower/. Anyone know of anything that surpasses this? Just making farming (perhaps one of the most labor intensive things you can do these days) less of a burden will prove to be a game changer (e.g. you may find a single person willing and able to maintain an entire one acre farm).

148 comments

[ 1.7 ms ] story [ 216 ms ] thread
I don't know what the maximum number of people you could sustain on an acre is, but my guess is way less than 1000. Like an order of magnitude less, perhaps closer to two orders less.

The world record corn yield is 500 bu/ac. 1 bushel of corn is about 1500 calories. EDIT: 1500*56=84,000 calories

If you could do 1000 people per acre, you would have enough food for 36 billion people from the state of Iowa.

A bushel of corn weighs 56 pounds (shelled) or 70 pounds (still on the ear). Using the shelled bushel in this case, 1 pound of corn = 1,656 calories so each bushel is roughly 92,000 calories. With the record yield of 500 bushels/acre that can feed 60+ people for a year!

I can't think of anything else that can come close to that kind of yield. No wonder corn syrup is in everything...

Whoops, forgot to multiply by 56 lbs in that
I think you're making an error there somewhere, possibly mixing calorie yields per year with human requirements per day.

Divide that by 365 for a more realistic estimate. About 60 people per acre if you can maintain the world record yield, or about 15 with the mean US yield.

500 bushels/acre is probably not scalable. For corn, if you get 150 bushels/acre, you are doing well.

I think the grand champion of caloric yield per unit area is actually the sweet potato. But potato is second place, has more protein, and a better sale price. Also, sweet potato contains raffinose, meaning that it will give you gas.

Browse fao.org for a bit for more information on the nutritional yields of various crops. Though last time I checked, there was no comparison with perennial tree crops like oil palm, neem, and jackfruit.

Really? I thought aquaponics and other advanced techniques drastically increased that by an order of magnitude. Doing a quick search, it seems that you're right. For an average person, with an average diet, to feed a SINGLE person you need an acre (with traditional farming techniques)[1]. Wow. I doubted a thousand people could realistically be fed on an acre, but I thought at least 10 would be trivial, and 100 possible with some efficient tech. Pretty eye opening, honestly (humbling moment, hah)

[1] http://www.farmlandlp.com/2012/01/one-acre-feeds-a-person/

EDIT:

The best I could find in terms of output in the US a farm that does about 40 tons with three acres, which is supposedly ~200 people [2]

[2] https://www.splendidtable.org/story/how-to-feed-10000-people...

Unfortunately aquaponics and vertical farming are often advertised as "the future of food" with bloated claims of yield increase. While controlled environment agriculture (CEA aka greenhouse farming) does have increased yields and an elongated production period, the gains are nowhere near the claims that are being made (factor of ten to factor of hundred). Apart from that greenhouse production is mainly used for produce and leafy greens. It does not make sense to produce grains in a greenhouse. Yet the bulk, like 90%, of our calories come from grains and leguminoses.

Plant yield without artificial lighting is limited by light influx. Prof. Bugbee from NASA co-authored a paper on this titled "Exploring the Limits of Crop Productivity : I. Photosynthetic Efficiency of Wheat in High Irradiance Environments" (DOI: 10.1104/pp.88.3.869). The gist of this paper is that one gram of dry biomass needs at least in Mol of (sun)light (for lettuce, other crops need even more light per dry mass).

It's unfortunate how much hype surrounds vertical farming and hydroponics as it'll likely cause a let down and potentially lead many people to miss the actual benefits. Vertical Ag won't completely replace large scale crop farming, especially given the record population levels we have nowadays. The "green revolution" of high yield mass crops has been a boon to world food stability.

Why the hype for vertical farming then? Variety, locality, and nutrition. While bulk calories come from grains which don't grow well in vertical system, many leafy greens and vegetables do. Importantly many nutrients (especially phytochemicals) degrade very quickly after harvesting. Transporting food like strawberries or leafy greens hundreds or thousands of miles results in quite severe reduction in nutrition, and often taste suffers and require significant refrigeration to counteract. [1, 3]

Great point about the lighting Influx! Indoor growing without greenhouses requires a lot of light. Alas, it's not well understood... We review a lot of lights for our customers and seem to regularly get companies who claim to increase power efficiencon over 90%! Amazing -- except they confuse lumens with light flux and won't get the corresponding plant growth. Photosynthesis requires a high light density, measured in Molson of light per volume. Kinda cool! Thanks for the link above, my coworkers will find it helpful.

All that being said we ended up developing our own custom LED lights using water cooling techniques similar to Google data centers. To achieve high enough light density while not frying the plants from waste heat, we pump the heat out via a chiller system. Basically, anytime you can water cool something -- do it! ;) [2]

1: http://www.fruitandvegetable.ucdavis.edu/files/197179.pdf 2: https://brightagrotech.com/coolbar/ 3: http://pure.au.dk/portal/files/45962257/PostharvestQualityCh...

(edits: grammar and forgot a link)

Totally agreed. Calories from the fields and acres and variety, vitamins and other nutrients from CEA production.

BTW: I'm following your BA media streams and I'm curious about the coolbar lights. Do you have distribution in Europe/Germany? Would like to get a quote for our (small) systems (NGO community and small research system, both aquaponic).

Great to hear! We do have a European distributor. He's a nice fellow over in France. Email me at 'jaremy @ BA', and I'll forward you to people who can put you in contact.

One thing to note on coolbar lights is that there's a minimum size due to requiring a chiller system. We're working on non-chiller based systems for very small setups.

Are you part of a research group? It'd be great to hear more about what your interests are and what tech might help. We're developing some interesting machine vision based tech too. :-)

I think the ballpark you are looking for would be Biosphere 2 [0]. This is a 3 acre enclosed system that supports the food and oxygen needs of 8 people, although the people lost a lot of weight and oxygen levels dropped dangerously. While this (somewhat failed) experiment is in the ballpark of what long-term spacecraft will need to sustain generations of crew, it also shows how much land is needed to supply the air we breathe.

Notes: 0 - https://en.wikipedia.org/wiki/Biosphere_2

The most land-efficient crop you can grow in the US is corn, at 15 million calories per acre. (In warmer climates sugar cane edges it out with about 17 million calories per acre). Your one-acre farm could feed about fifteen people, not one thousand, and that's if people could live on corn alone.

If you want to improve agricultural technology I'd suggest finding and talking to some farmers and finding out what they want, rather than starting from a clean sheet and an idea of what a science fictiony farm ought to look like.

>an entire one acre farm

The way you wrote this reads as if you consider one acre to be a very large tract of land.

I have family that farm and I can tell you that one acre is actually quite petite as farms go. The average size of a farm in North America is something like 400 acres.

One of the farms I run has a small 2.5 acre section of the main farm divided by a ditch. If that chunk were much smaller, I'm not sure it would be even worth pulling the equipment into it. If it were any smaller, some of these large implements you can buy nowadays wouldn't even fit.

But perhaps what he means is that with more automation, the new tools would make utilizing small pieces of land up to 1 acre more attractive. I'm not so sure, but there's always the idea thrown around of automation allowing equipment to get smaller (and large operations would buy many of them).

Indeed. What I imagined was that if a single acre could produce a high yield, balanced diet, then they could simply be placed in places where it before wouldn't make sense: e.g. interspersed within the suburban sprawl. And since it's just an acre, a few people or a single person could take care of it, no heavy machinery required.
But an acre of land in the suburbs costs hundreds of thousands or millions of dollars, whereas an acre of land in a normal agricultural area far from a city costs a few thousand dollars.
maybe there has to be a certain chunk of land set aside in urban areas for green and open spaces? concentrating massive food production centers is a terrible idea. not just because its an easy target for bio hacking but also because of the impact on environment and water especially. small distributed food production centers makes much more sense.
In the US, 40% of all its land is in active agriculture production. Most of the remaining 60% is not in production because it is not productive. I suggest production has already spread to everywhere that is viable to do so.

Even if you can find and allocate several thousand acres of productive farmland in every city for growing food, what are you really accomplishing? That's but a tiny little drop in the bucket of all the land needed to sustain the population. Not to mention that many cities rose out of agrarian pasts, so it is quite likely that active farmland is already found on the outskirts.

i agree. its not worth it for me to buy large equipment to do just one kind of farm task. however, smaller machines that do a variety of task with minimal supervision will be immensely useful. large farms grow commodity crops. small farms(a 100 acre farm is considered a 'small farm')grow cash crops/food we eat.
I was looking into this a while ago, i do remember there being a farm bot being kick started(1). I was wondering about doing something like that myself, but for a hydro solution for indoors - i am about to start at a new restaurant, so that has to go on hold at the moment though :)

Note: That farmbot i linked is open sourced, so i will personally be looking more into that when i have more free time.

(1) https://farmbot.io/ (seems to be about to ship the first batch)

The fully automated farm is a few years off. Here in Michigan large families with one or two employees farm 5-10,000 acres.

With good Iowa or Illinois farmland you can expect 200 bushels of corn per acre or 60 bushels of soybeans. This land by the way will cost you approximately $7,183 per acre.

If you're interested in this topic you might consider attending the precision ag conference, the next one will be in 2018

http://www.precisionagvision.com/

Or your vegan community could move to downtown Detroit. Cheap fixer upper houses going for $1,000 with plenty of inexpensive land nearby. City brings the water right to you and with drip irrigation you can water your crop inexpensively. Lots of restaurants that want your produce and the largest farmers market in the country. http://www.easternmarket.com/

Lots of inexpensive, though untrained, help for minimum wage. Wide lightly travelled city streets and they just rewired the entire city with LED streetlights. Five million people live in the greater Detroit metropolitan area. You will have be proactive on crime, lots of web cams would be a wise investment. It wouldn't hurt to be armed, in fact the police chief recommends it! But without risks there wouldn't be the immense opportunities.

> This land by the way will cost you approximately $7,183 per acre.

That's surprisingly cheap. A farm near me, with similar production capabilities, recently sold for $25,000/acre. Around here, you won't find anything really worth farming for much under $15,000/acre on the low end these days.

That's an average price of Iowa farmland statewide. But as you know there's a wide variety. I have no problem believing the price of $10,000 to $15,000 per acre for the premier land.

However as you probably are aware sometimes competition among neighbors boosts those prices into the stratosphere.

Then there's Detroit where you can buy city lots for $50. Assuming an average size of 50 by 100 ft or 5000 sq feet that is $435 an acre and it comes with water!

http://auctions.buildingdetroit.org/sidelots

Average land price is similar here, but you won't be producing anywhere near 200bu of corn or 60bu of soys on that average land. Maybe 2/3s of that, if you get really lucky, but more like half in a more typical case for average and below. So, I still maintain that it is quite cheap.
That $50 also comes with obligations to pay back property taxes, renovate existing buildings to meet current code standards, and improve the street-visible appearance of the lot. You're paying $50 for a property that has a $50000 debt stapled to it.

Also, urban soil frequently has chemical contamination. If nothing else, it often has high levels of lead contamination from mere proximity to the streets used by cars that burned leaded gasoline before it was banned.

Buying multiple distressed urban lots to use as farmland is a horrible, horrible idea. If you want to buy land for traditional farm, do it in a rural area. Urban land does not make sense unless you are experimenting with vertical farming, where the grid power and piped potable water is already available, and the local soil (if any) is not going to be used anyway.

You are wrong. If you're buying a house it might have taxes attached but not side lots if you read the site I referenced. The city is literally giving side lots away so that they are added back to the tax rolls.

Also absolutely no one is going to buy a house with $50,000 of taxes attached! Instead you wait until the land bank owns it and either buy it directly from them or wait for the yearly auction.

Rarely do housing sites have chemical contamination. There are however former industrial sites that do. But if you look at who is doing urban farming in Detroit it is exclusively on former residential land.

Urban farming does make sense economically. Lots of young people don't have the capital to buy good rural farmland. If you're in the city you're also near great markets but the key is growing high value, labor intensive crops like fruits and vegetables.

Unless you're an experienced farmer operating at scale you can almost always buy corn, wheat and soybeans cheaper from the farmer than you can grow them yourself.

I encourage you to explore the difference between property taxes within the Detroit metropolis and in rural Michigan. Whether the back taxes have been forgiven or not does not obviate the future tax bills to the new owner.

The $50000 bill is the average code to bring buildings up to code. You are correct that this would not apply for side lots. But people can and have purchased houses for "$50" that included a ~$50000 obligation, for use as low-income rental properties. Such houses can be made profitable by leaning heavily on the federal Section 8 voucher program. And that's why those properties sell.

As for lead contamination in urban soil...

http://modernfarmer.com/2014/07/lead-urban-gardens/

https://soils.org/discover-soils/soils-in-the-city/soil-cont...

Urban gardening only makes sense for the most lucrative of cash crops, and those intended for consumption by rich humans. You could certainly feed yourself and your family with a well-managed plot, but you're going to have some trouble selling to anyone else, unless you serve an incredibly specific niche and have some loyal friends in the restaurant industry. The business simply doesn't scale. After the restaurant buys all it needs, you're back to the farmer's market, where even the top-shelf local CSA group can match your quality and your price, at 100 times your volume. Your only advantage is that you can deliver fragile-but-unbruised produce by hand with a bicycle fifteen minutes after harvest, rather than after an hour on a delivery truck. But in practice, hardly anyone needs that enough to pay the extra cost for it.

The guy growing fingerling potatoes on a 10 acre plot just past the suburbs is going to pay off their loan and then enjoy the profits, as the urban gardener works their ass off barely doing better than break-even.

Where do you live? Prime land in California? Average acre in America is $4000 [1], so wouldn't call $7183 an acre cheap.

1: http://www.agweb.com/mobile/land/farmland-value-guide/

> Average acre in America is $4000 [1], so wouldn't call $7183 an acre cheap.

But presumably we're talking about land that specifically can produce the 200bu of corn and 60bu of soybeans the the parent specified? There is most certainly cheaper land around, even locally, if you are willing to accept less production ability, but there is good reason why that land is cheaper.

The average corn yield in the US is 160 bu/acre and 49 bu/acre for soybeans, so it seems that the average land is not capable of delivering. Not to mention that there is land that wouldn't even be worth trying to grow corn and soybeans on, which would drag the average down even further if you tried.

That average has to include vast areas of aggressively useless and unwanted land (eg Wyoming).
Wyoming's a great place to start a wind farm.
Yeah, although Wyoming land is reportedly good for some things (eg grazing), quick Google searches seem to suggest price per acre more around $450 an acre statewide (http://trib.com/business/land-values-skyrocket-in-parts-of-w...).

The thing is, agriculture fertility is not the only factor in rural land price these days -- I've heard ala the linked article that investing in land for hedging / portfolio diversity purposes is popular. It's also no surprise to hear that more aesthetically attractive rural areas will fetch a premium (I would seriously doubt the area around the Grand Tetons region is just $450 an acre for instance; the linked article suggests as much).

Do I have to be a US citizen to buy land in US?
I don't think so. There are rich Chinese people just buying housing property in Arcadia, California and Vancover Canada... I would like to think housing property and land are the same thing.
No, although you'll probably want a lawyer in addition to a real estate agent if you're a non-resident, and financing might be trickier as well.
Those are some pretty wild corn yields - 5 ton/acre - is it irrigated?

In Australia dryland farming about half as productive is around $2,000/acre

Iowa is mostly not irrigated.
Across the heart corn belt (iowa/illinois/indiana) we see 200+ bu/ac yields for corn consistently and majority of fields are rain-fed (non-irrigated). Hell in 2014/2015 some people around me (champaign IL) were hitting 220/230. Illinois has some of the most productive soils in the world though, we are lucky!
That nice black Illinois topsoil. Some good stuff.
(comment deleted)
You see yields like that in the US west of the 100th meridian. It's a stark divide. On the east side, farmers put seeds in the ground, rain does the rest, and you get amazing yields. West is substantially more dry. There, you see irigation and dry land techniques.
Check out https://farmbot.io which is a smart automated bot for raised beds.

Also read up on permaculture which attempts to create better efficiencies through smart design.

There was a great BBC doc a few years ago (2009) called "A Farm for the Future". It was about someone from a traditional farming family discovering/researching/interviewing around permaculture. In one of the interviews she mentions that traditionally she can feed 5 people an acre, whilst the interviewee was claiming 10+. I believe permaculture has come along in leaps and bounds since, so it is well worth reading up on.
> Do we have any technology (independent of cost) existing today to have a fully automated farm?

From a grain farmer's perspective, the modern equipment you get nowadays is practically fully automated already. However, that's the fun part of the job! While I do get excited about the automation, I am also in no huge rush to take away the human touch completely. After all, I farm because I want to be out in the field.

But keep in mind that the field work is also just a small part of the farm operation. Much of the work happens back in the office. I don't see that being automated any time soon, at least no more than any other business could be completely automated.

Regarding your second question, you imply that it would be easier to automate the vegetable production than to automate livestock production. In fact, current dairy farms are in many ways more automated than crop farms.

Modern dairy farms feed, milk, vaccinate, etc. automatically--without any human intervention. The automated systems alert humans if an animal does not show up for food after some threshhold of time. Otherwise, it can even supply custom rations per animal, detect various ailments, track weight gain, monitor milk quality all by itself.

For the more general gist of your question--complete self-sufficiency on a single acre, that is more problematic to automate. Industrial farmers already automate a lot. They have automated machinery to re-level fields, plow fields, plant fields, and harvest fields, and irrigate. but those systems are largely practical only for large parcels. They operate on fields of, for example, 40 acres. And they are sufficiently expensive as to require many hundreds of acres to justify their cost. Also, they tend to be single purpose. They can only harvest corn, or only wheat. Planting a variety of crops within an acre would require a variety of automated tools. Think how different your equipment would have to be to harvest carrots as compared to barley.

I don't mean to say that it can't be done. But it would take a lot of work from where we are now.

Incidentally, your seedstock.com link is not impressive because it is "advanced." It's most impressive feature is its size. Hydroponic greenhouses tend to be more labor intensive (but have higher yields), although you could put the same attention into earth-grown crops if you chose.

There are lots of interesting agriculture use case being enabled by drones. "Ground" robots are still too expensive and unreliable, so they are not widely used yet. On the other hand, there are estimates claiming that 5-10% of farmers are using drones[1]. In addition to providing analytics, some drones already perform active tasks. E.g. take a look at spraying drone from DJI[2]. As another example, there is a startup that produces a drone that shoots weed in your field with the laser ;).

1. http://bestdroneforthejob.com/drone-buying-guides/agricultur...

2. https://www.dji.com/mg-1

Shameless plug: I am a co-founder of the startup working on deep learning from drone imagery - http://tensorflight.com. We primarily work with agriculture. Please get in touch!

What are your goals for you research or project? Many ideas stumble not on the technology, but on the people aspects of the project.

If your goal is a single family sustainable farm, I've read reports of 2 or 3 acres being plenty for a family of four if you are in an area with enough rainfall (Ohio, for example). However, they needed to have outside income since their land did not produce everything the needed.

If you are thinking of a larger group like commune of 25 people, or a town like you mention, you will to look into the issues present in those communities, and getting food is usually not the biggest issue. For example, several co-housing projects fail due to the families involved not being able to raise enough money to purchase the land the need. Other small groups buy farmland but don't have the capital or expertise to actually farm and produce enough to be economically viable. Useful farmland is not cheap, and the economics get in the way quickly.

Farming is a complex business. It won't be "fully automated" anytime soon, particularly for the variety of crops people like to have. For example, my parent's potato farm also had a vegetable garden for carrots and onions, and they raised chickens. They ate a lot of soup, since that is what they grew without needing to spend money, meat was expensive. Economically they sold potatoes to have money to buy everything things they needed (gas for tractor, salt, tools, etc.).

Note: I am a cofounder at www.powwowenergy.com, a data mining company based in San Mateo CA where we help farmers be more efficient. If you are a data scientist or developer looking to help the world feed more people, drop me a note, we are hiring.

This is just a tiny part of a much broader goal, which is to create a completely self sustaining town of 100 to 1000 people within my lifetime. My intention is with 100k that could get you 10 acres. 1 acre for food, 5 for solar/energy production and 4 for housing/community. Of course, with the comments in this thread, I'd need more land, haha.

As far as the economics goes: I've been doing reading but haven't formed a good conclusion yet. Initially I thought a communalist approach would work. I thought of too many ways that would fail and then opted more for a centralized, socialist approach, but that seems like it would fail as well. Capitalism, of course, would just result in what we have now, defeating the purpose. Personally I think communities where each person knows everyone else have unique advantages not present in large cities.

Disclaimer: I haven't done a ton of due diligence on these things yet. When I get closer to actually doing this I'll start doing more serious research.

This may be exactly the opposite way to look at it, but have a peek at [1] the One Straw Revolution. TL;DR: Masanobu Fukuoka was a scientist with an idea of using existing biological systems to reduce the amount of labour involved in farming. He created a very nearly zero-input farming system with the hope of enabling people to grow enough food to sustain themselves in their spare time.

Unfortunately, lots and lots of hype surrounded the project and reliable numbers of yield, etc, etc are probably not to be found. Still there have been lots and lots of similar initiatives with interesting results. I think you would need quite a lot of land for 100 people, but land is cheaper than robots (and is self repairing, and adds to your quality of life). I have a friend who grows all his rice and vegetables for the year in his spare time using Fukuoka's methods (he studied with him for a while). I helped him harvest once and I was quite surprised how well he was doing (easily as much yield as his neighbour's commercial farm). Farm land is trivial to get in Japan and you can often borrow enough to live off of in exchange for a bottle of whiskey or so per year.

[1] http://www.onestrawrevolution.net/One_Straw_Revolution/One-S...

How will this town get medicine and technologies like replacement solar panels? Will people pay with USD into a general fund that would be used to buy from the outside world? What happens if people can't pay?

Sorry about all the questions, but a community that small can hardly be self sustaining in a real sense without lots of imports or a massive step down in quality of life.

Indeed. Ideally (need to do more research) the community would self sustaining with a low quality of life, however, all of the usual amenities would be provided. The difficulty is setting it up so that they don't necessarily become entitled or dependent on the advanced amenities.
The keyword here is "intentional community", and there's a long history of people trying to make it work and sometimes succeeding nicely.

But neither farming nor technology is key to making it work. The challenges in decreasing order of size are: interpersonal (requires a lot more collaborative effort and management of drama than you'd expect); legal/administrative (you still need to pay e.g. property tax & comply with building codes & farming law); cashflow (you still need to buy from outside, especially if you're hightech); and only then the day to day of growing food (which may be more labour than you or the community is used to).

On the <100 people scale it's like having roommates who are also co-workers. Hence the tendency for intentional communities to come with some level of cultishness to ensure the community sticks together without being pulled apart by individuality.

As a wine grape farmer I have to say that there is not much going on.

There are people using drones, but I don't see much actionable data going out of them (especialy as you check each individual plan few times a week anyway). There are reports about some robots being developed, but nothing much being deployed.

State of art technology is optical sorting machine, which check each berry, after being destemed, and discard those which are unripe, molded or damaged. It is kind of cool and I hope to be able to afford it some day.

Isn't there at least some buzz around optimizing water usage? Fine grained monitoring of individual plants using cheap sensors for example.
It might be case in California: grapes are ok without any irrigation, unless you are farming in couple of really hot places, or you really want to boost your quantity instead of quality.
Relative to HN I'm probably an expert in ag. Relative to people who grow things I'm far from an expert. I was employee 3 at Blue River Technology and I've been there for four years: http://www.bluerivert.com

The way the Valley generally thinks of ag is completely broken and I highly recommend avoiding Valley people if you want to learn about it. "Fully automated" is a thing people who've never farmed before assume is possible and assume people who farm want.

There is a lot of value on the table in ag tech. Depending on what your objectives are, I recommend having as many face to face / dirty boots meetings w/ growers as you can. Understand the realities and nuance of their day to day and find a place where you can make their lives better. Ignore the sensational articles. Most "ag-tech" companies are comically detached from the realities of their intended customers.

If your goal is to grow things then follow the above advice w/ 10x the emphasis. Happy to help further. Contact in profile.

Why is fully automated agriculture impossible?
Impossible may not be the right word, but highly unlikely due to the environment ag automation needs to operate in. Production agriculture is highly mechanical, meaning things wear out and break all the time, which means humans need to go and fix things. In addition, people farm in every environment on earth, with wide variance in conditions in even the most ideal places. This means mechanical equipment needs to be specialized for the crop or location (considering humidity, temperature, wind, sun, crop type, etc), which makes scalability limited. These are the two biggest issues that tech doesn't really have a good solution for.
I'll take a quick stab. First off it's not a software computer system. Their are so many different elements in play. Your robot needs to know about soil science, agronomy, pests, weather predictions, stock markets, machinery, the list goes on.

Farmers have to wear many different hats. You are a seed scientists for part of year when you figure out which seed you should plant for next season. This choice involves many other factors (weather predictions, personal yield goals etc). During planting season you have to make human decisions of when to plant/not plant. Training a robot to make this decision would be very complicated and most likely involve many 'failed' attempts that cost tons of $$$.

Lets say you did get the robot to choose a good seed, then you hired a fleet of robots to plant corn on 5000 acres and they successfully chose the correct times, right plant depths, row spacing, seeding rates, and managed to get the crop fully planted without any hiccups.. At this point you would then need to be scouting the fields, most likely putting down some N after planting. Robots have to decide which fertilizer goes where, how much to put on, is each field the same or does field X get higher N on one side but less N on other??

THEN lets assume all your crops grow fine, weather behaves (never does)and your robot fleet has scouted all fields. Then the robots must go out and harvest the grain. The combine picks up grain which then gets dumped into a semi on side of the road. Again maybe there is a 'harvest semi for Uber' kind of service where automated semis drive around between combine and elevator at will.

Another decision to be made is selling the grain or storing it...

In short, there are way too many decisions and farmers (i'm thinking midwest corn/beans guys) have to make across many disciplines. These guys have an intrinsic knowledge about their land that has been accumulating for 10s or 100s of years!

Growing up on a family farm & surrounded by farmers, most of them are not great at any of that. I do think there is a lot of potential for automation for each one of the things you mentioned. At first it can help farmers make better decisions. Slowly it can grow to make those decisions on its own.

A 'huge' part farmers struggle with is trying to guess when to purchase & sell their products. That's everything from the end product to the seed & fertilizers. Also, tech could help farmers get a heads up on if there might be global surplus or shortage on certain crops. Sensor based tools can do a much better job on how much & where to put fertilizer. I've even seen demos of not using pesticides but friggen laser beams to shoot only certain types of insects.

My prediction is future farms will look like giant JavaScript projects with crazy long package.json files!! There are so many possible tools that could improve a farmer's life. I can even see "Tool Fatigue" being discussed at the local gas & coffee shops.

We are working on a sort of "AWS for farming" concept here (http://www.habitat84.com/). The goal is not to make fully automated robot controlled farms, but rather to reach an appropriate level of abstraction such that farming systems can be controlled and scaled easily by users (who might be farmers, but also others). The goal is to develop a platform on which to ultimately solve the problems of food production in space.
Agriculture in our age must shift from productivity first to sustainability first, because we are running out of fossil energy allowing to produce most of the fertilizers, and we have already killed our soil with it. A farm involving a lot of machines (i.e. a lot of need for energy) might not be the way to go in order to keep the agricultural system stable. The most you ask from the soil, the poorer it gets, and then you have to dump fertilizer to keep some level productivity.

We have created flawed models of what was good for productivity. I work with a guy very invested in research on olive growing (here in the south of France), he's achieving results believed to be impossible by many agricultural researchers. His method is entirely organic, shifted towards correct use of pollination, cross breeding and a lot of care, the kind of which automated sensors remain lacking of. From what he told me, he's able to develop the root system of an olive tree ~4 times as fast as researchers using what is believed to be optimal conditions.

We already have excellent robots able to take care of farms, they're called humans. Farming is a very noble activity and instead of less farmers, we need more, with more time to study the impact of pollinizers on their crops, how to deal with the new pollinizers they'll get due to climate change, etc. All of which can't be done unless we re-humanize farm work.

This comment is spot on. There's good reason to believe that the low-hanging fruits in improving agriculture are non-tech changes in farming. For example, a province in India was able to improve rice yields by 45% by transplanting seedlings earlier in a grid pattern [1].

I doubt that eliminating human labor costs would move the needle, automating farms seems like a low ROI move.

If you follow the money, it's easy to see why the US has focused its research on high-tech petrochemical fertilizer farming and GMOs. You can sell proprietary fertilizers, herbicides, insecticides, and seeds, but it's much harder to monetize new farming techniques.

[1] https://www.theguardian.com/global-development/2013/feb/16/i...

>I doubt that eliminating human labor costs would move the needle, automating farms seems like a low ROI move

If that were true, the farms in California wouldn't be filled with illegal immigrants.

> If you follow the money, it's easy to see why the US has focused its research on high-tech petrochemical fertilizer farming and GMOs. You can sell proprietary fertilizers, herbicides, insecticides, and seeds, but it's much harder to monetize new farming techniques.

I believe this means that you are not aware of the work of Norman Borlaug [1] [2]. Basically the very fact that we don't have famines now in places like India, Pakistan and Mexico is due to his and his group's work on intensive agriculture, which was mostly about farming techniques, government practices and setting up local fertilizer productions.

Moreover, it seems that you somehow look down to fertilizers. There is nothing wrong or inherently "non-green" in fixing nitrogen into nitrate fertilizers, you can easily do this with solar energy.

[1]: https://en.wikipedia.org/wiki/Norman_Borlaug

[2]: https://en.wikipedia.org/wiki/Green_Revolution

Right, I've heard the Western narrative on Norman Borlaug and the "Green Revolution".

I'd highly encourage you to read through the work of Vandana Shiva. [1] She'll provide a well-researched, critical view as an actual stakeholder in India's food system.

I look down upon petro-chemical fertilizers because of their disastrous effects on the environment and food security. Happy to elaborate, but Dr. Shiva work does it better.

[1] http://vandanashiva.com/?p=291

You keep throwing the phrase "petro-chemical fertilizers" around but I can't see how it can make sense except for denigrating the concept and spreading FUD (which is counterproductive). Ammonia is just NH3 and there is nothing "petro" (i.e. hydrocarbons) in it, that hydrogen can be obtained anywhere and the nitrogen is literally around you.

Can you please elaborate on how fixing nitrogen on industrial scale is "disastrous"? What difference do you see between bio-available nitrogen that originated in root nodules and chemical reactor?

I just listened to a podcast (50 things that made the modern economy http://www.bbc.co.uk/programmes/p04f77rg) about the Haber-Bosch process https://en.wikipedia.org/wiki/Haber_process

According to the podcast and the wiki page, the current source of hydrogen for ammonia is natural gas. Additionally the process is energy-intensive, using high temperature and pressure.

These things don't necessarily make it mandatory that it's petro-chemical, but due to the way we currently run it it is.

And the current source of electricity that runs HN is most probably fossil fuels. Does it make HN "petro-website"? It's clearly a derogatory term that serves no constructive purpose. I believe we should strive for zero-emission hydrogen production (which is clearly possible given that hydrogen can be easily used to utilize excessive production from "green" sources) instead of pushing luddite worldview.
The key point of my post was "the current source of hydrogen for ammonia is natural gas" It can be made from other sources, but it is not. Hence petro-chemical. Like plastic or nylon.
> You keep throwing the phrase "petro-chemical fertilizers"...

Given that most fertilizers are composed of chemicals derived from petroleum, it's accurate. I don't use the term "conventional farming" because these techniques are extremely modern.

> Can you please elaborate on how fixing nitrogen on industrial scale is "disastrous"

To be clear, fixing nitrogen is fine, bacteria do this already naturally. The problem with fertilizers used at the industrial scale is that they end up ruining local water supplies + killing local ecosystem of bacteria and insects [1]. After a few years of using industrial fertilizers, famers end up ruining their soil and end up being dependent on fertilizer suppliers for future crops.

[1] https://www.scientificamerican.com/article/how-fertilizers-h...

> most fertilizers are composed of chemicals derived from petroleum

Can you please elaborate further? Yep, in a lot of cases hydrogen for ammonia comes from natural gas. Nope, it doesn't have to be, in fact you can easily get it from water. What other "chemicals" do you have in mind?

> because these techniques are extremely modern

They are, but the concept of "not dying of famine" is also pretty modern if you consider an overall population.

> The problem with fertilizers… killing local ecosystem of bacteria and insects

I believe you confuse fertilizers with pesticides.

I feel like you're not answering the question and instead subverting it to discuss your own philosophy. As legit as it is I don't see how this is helpful to the person asking the question.
I am trying to provide context regarding the whole problem. Besides, the guy I talked about uses heavy technology whenever he can (he needs special equipment to take macro pictures of bugs, he needs social interaction with other people in local vicinity spotting larvae of a pest, he needs way to statistically test land lots, he needs ways to measure sun exposition and time, etc.) ; the last thing he needs is robots doing his job though.

And every question is as loaded with philosophy as any of its answers. :)

>I am trying to provide context regarding the whole problem.

You're mistaking context as an opportunity to share your personal views and beliefs. OP asked very specific questions. Your view on renewable energy isn't part of those questions.

He's pointing out flaws underlying the question. Definitely relevant.
Sometimes, people asking questions are actually asking the wrong questions. This is often true. So responding to the question with a better question to consider is a perfectly valid answer.

In my experience, if I don't know the answer, it's usually because I don't understand the question.

I don't see any justification though for why humans need to do the labor. This guy found a nice pollination technique. Great, teach it to the robots or spend time teaching the robots how to analyze different techniques for effectiveness.

Humans are not great at farming, that's why we have entire industries producing tools to help them do better. It's silly to think that riding around in a tractor all day is a better long term solution than checking up on a robot doing it.

The reality is that you are not going to feed a town with a tractor or a robot acting as a tractor. This is because with your robot, you are producing cereals. If you want to feed the town, you need fruits and vegetables.

Now, if you want high density, sustainable fruits and vegetables production, you need to pay extremely attention to your soil. Practically it means, try to never walk on it and produce multiple species on the same area to use the different patterns of root networks, shadow and cross protection effects.

If you do that, you end up being highly productive on a small surface but it is also very hard to see opportunities in using robots. Take a look at a very smart people running a small farm[0], he is relentless in using and building the right tools to save work and improve output etc.

[0]: https://www.amazon.com/Market-Gardener-Successful-Handbook-S...

People can survive on cereals.
not cereals alone, and not if they destroy the soil in a few generations
And primitive hunter-gatherer people tend to be taller, more intelligent and healthier than primitive agricultural people.
The only way you can even come close to automating farming is through massive conventional monoculture systems. These systems destroy the soil and sacrifice genetic diversity to achieve this goal while producing a glut of a single crop that ultimately drives the value of that crop into the ground.

Humans are incredibly efficient - our pattern recognition ability on a per watt basis is way beyond what we can do with machines. Not to mention we handle environmental irregularities far better than most robots, and we have far more task flexibility. We have millions of years of evolution behind us to make us efficient foragers, it is silly to think we're going to beat that with a few hundred years of science behind us. Instead, we need to figure out how we can make agriculture an enjoyable career instead of soul crushing toil.

Agriculture used to be the dominant industry in my part of Canada, and I have relatives/many friends who either farmed or currently farm. Consequently, I was once extremely interested in agtech.

I ran into a couple of problems. One big problem is that farming is not a particularly good business to get into right now. Equipment costs are very high, inputs are constant, and the global market is only kind to those producers who produce at massive scale. The days of taking over your Dad's quarter section and feeding your family ended before I was born. Now, if you want to take over the family farm, you more likely than not have a full-time job in the nearest community...

Because of the economic crunch, it has created a demographic issue. Many of the farmers I spoke to are in their 70s and 80s. They are mainly retired, but keep working the land to keep it away from corporate farms. Farming is more of a hobby at this point, and they choose to do it the way they always did.

The young farmers who would be early adopters hesitate because they more often than not rely upon their parents and grandparents during seeding and harvest. And, they can never rely upon a profit big enough to even pay the cost of automation.

Edit - It occurs to me that the phrase 'quarter section' has little meaning to anyone outside of agrarian communities. A section is usually one square mile. In Canada, in the late 1800s/early 1900s, the government granted one quarter section to any family that showed they could work the land. The Dominion Land Grant fuelled immigration. Many families still own their original quarter section and are fiercely protective of it.

The justification is that we're in an artificial and unsustainable situation at present. Just 8 generations ago, over 90% of the labor force was farming. We're at 1% now. We came up with some incredible engineering practices to achieve this but we lost sight of sustainability in the process. It's the most fundamental good in the economy after water and before shelter. It's reasonable that more than 1 in 100 people should be involved in producing food.
Sorry, but I think this is nonsense.

Robots are incredible more efficient and energy saving than humans.

Do you really thing a human would use less energy driving a car/tractor than a robot?

Robots never lose attention, are very easy to measure their performance and are incredibly accurate.

And I'm not sure why humans would use less fertilizer/pesticides than robots?

If you have a system that works better, then tell the robot and it will do it.

Or are we tilling by hand and introducing all the work place deaths this introduces? Farmers die at a high rate, I don't think it's that noble in it's old form.

For simple, monotonous tasks, robots are indeed far superior to humans.

For more complex tasks that involve flexibility and high level pattern recognition, humans as so far above robots on a work/watt basis it is silly. We have been developed by millions of years of evolution as efficient foragers.

Of course, if you take the complexity and pattern recognition out of farming you make it amenable to machines. An unfortunate side effect of this is that the farming practice becomes unsustainable.

Robots can do repetitive tasks in tightly constrained conditions faster and more accurately than a human. But anything that involves incomplete/inaccurate data, wildly varying conditions, and most importantly, decision making is best left to humans.

That's why computers run code, and humans write code.

The most important fertilizer is some form of ammonia (bioavailable nitrogen) and you can't "run out" of nitrogen really, not on this planet. Energy requirements of fixing nitrogen are high, but the process is quite robust to fluctuating power supply, so you can easily use "green" energy to do this. So in the end you either let bacteria do this (e.g. in legumes' root nodules with questionable energy efficiency) or you do this on industrial scale, using solar energy to efficiently fix nitrogen and introducing it manually to the soil. It should not be a question of "what's natural" or what feels better, rather it's a question of energy efficiency and reasoning from first principles.
Is these a way to get in touch with this person, follow their work, read their research, etc.?
> Agriculture in our age must shift from productivity first to sustainability first

But the two are not mutually exclusive; productivity has a role to play in reducing the footprint of agricultural activity. I don't see it as a dichotomy of 'different OR less', but 'different AND less'. I don't think we should make the mistake of inferring that productivity necessarily means greater and unsustainable use of energy and natural resources.

Only in this age of low interest rates does high productivity equal unsustainable use of resources, as the price of anything with a yield (like arable land) becomes prohibitively high for a farmer to have grazing cattle, such that buying a huge barn and putting the cattle in there is at least half the cost (that's what's reflected by the supermarket prices I see, at least). This changes as the rate of interest rises.

With a falling rate of interest, it becomes increasingly cheaper to borrow capital to buy machinery and build barns, and put live animals in these, while any asset with a yield -- such as farm land, or bonds -- is increasing in price.

With a rising rate of interest it's the opposite. Borrowing money to rent machinery and build industrial real estate becomes increasingly expensive, but the price of land will be falling, as there's less and less reason to purchase farm land, in order to earn a yield, when the short term rate of interest is rising (basically risk-free profit).

> Instead of less farmers, we need more

That's bad for the economy and productivity, and for the environment in the end because more farmers with smaller land are squeezing for gains. It's especially a problem in in the EU where agriculture is heavily regulated and EU-subsidized. Take a look at the countries with the largest agricultural population: https://en.wikipedia.org/wiki/European_countries_by_employme...

And all this while keeping in mind that the common agricultural policy in the EU keeps the prices high for european producers.

But if you follow the parent's argument, food prices are artificially low because they don't factor in fossil fuel limits/emissions as well as other effects of soil depletion.

One big problem in America that we haven't even begun to fully address is the Ogallala Aquifer's rapid depletion. Without it we will need truly massive infrastructure investments to maintain one of the world's biggest breadbaskets.

So maybe more farmers would be bad for the economy now, but eventually we are gonna have to pay the bill for all the quick and dirty growth we've built our society on, which means things could be much, much worse for our parents or grandchildren.

Oh god. This times a million. Big Agra is just slamming in N-P-K 10/10/10 and using center-pivot irrigation[0] during peak sunshine hours because hey, the water table's only 15 meters down! Then to stay competitive average farmers have to beat their land just as hard.

Here's a guy who runs a mini-vineyard in CA who saved almost half a million gallons of water in one season for his grapes using pretty trivial means[1]. I bet if he incorporated some recent 'best/alternative' practices he could get that significantly lower too.

The good news is universities (particularly Australian unis[2]) have conducted pretty thorough analysis w/r/t efficient watering for healthy plant root systems. You have lay new sub-surface irrigation drip system with textile enclosures[3] which is an initial expenditure, but you end up saving a boatload of money on water in the long run. (Drip irrigation[5] in general is alright, but you have tons of water loss via gravity[4] as shown here.) "Permaculture"[6] is a DIY-sociological movement of farming by 'efficient/sustainable/whatever' means.

The bad news is the last time the US had a radical shift in farming was when farmers beat their land so hard we ended up having the Dust Bowl. Conditions had to get so bad that you'd get 'dust days' like 'snow days' from school/work for people to change their farming habits. It wasn't until FDR appointed Henry Wallace(unsung national hero) as Secretary of Agriculture that things started to change. If I could buy futures in water 50 years from now, I'd put my entire savings in it. The botanists and horticulture scientists have made massive advances since the 1930s. We can sustainably farm on land every season if you don't plant corn every dang season and then pump nitrogen in there to make up for the lack of nutrients.

Grey water[7] can very easily be treated and reintroduced into your local garden. We can even use refuse from septic tanks, treat it on-site and re-introduce into your your lawn/garden to produce 100% safe edibles (subject to your local regulations; also, if you're on any sort of medication such as antibiotics, you don't want to introduce that into your ecosystem). No land? No problem. You still can save water. Get a dual-flush toilet[8] which uses half the amount of water to dispose of urine.

I'm not a 'save the earth' kind of guy. The planet was here long before humans and will still be around regardless of all disasters we can possibly throw at it; that's a certainty. Will your grand-children have congenital defects from rampant pollutants, poor air quality, and tainted water is the real question. (I'll end this on a good note - the good news is reforestation can be done in a generation or two![9])

--

[0] http://www.lindsay.com/stuff/contentmgr/files/0/6ab577c195d6...

[1] http://hackaday.com/2015/12/11/vinduino-full-irrigation-with...

[2] http://www.unisa.edu.au/Research/CWMR/Students-and-researche... - These guys are doing some interesting work. Most of the Aussie unis are actually. U of Kansas is doing some great work too.

[3] https://en.wikipedia.org/wiki/Subsurface_textile_irrigation - They mention PE and PP but there are alternative materials availab...

Where do you think all the water from your septic tank goes? Right back into the water table, and back up your well pump to drink again.
You are correct, except aquifers like the Ogallala don't replenish in years or even decades. It takes millennia.
Do you have a link to this olive farm you speak of? Like a facebook page or something?
I have lately been exploring agricultural technology as I plan to start farming myself. Here is what I have learnt so far.

1. Yes, its possible but its not as simple as ordering stuff off Amazon. The farmhack.org community is a great place to start. Most of them are existing tools which have been redesigned by farm hackers. http://blog.regehr.org/archives/1446 is an interesting proposition too. But in my interviews with farmers from rural India they said, cost/benefit isn't much for the such aerial reports.

2. One Million Dollars!!! Thats what Jean-Martin Fortier says in his book; The Market Gardener: A Successful Grower's Handbook for Small-Scale Organic Farming. Its a great read as its not just about tech but also market.

For robotics, try the search term "precision agriculture". This application of robotics is not so much about using the robots to replace manual labor, but rather to gather more detailed information about crops. An example application is yield prediction, which is really valuable as a predictor of the amount of laborers needed to harvest the crop within the window of time opportunity.
All I know of is Clean Seed (TSX.V: CSX) which is a company that produces an "intelligent" planter, you can set it up to vary the depth and seed type planted depending on your map of your fields and varying soil conditions. They're trialling this year and the big question will soon be answered about how much it impacts yields.

Its in that direction but obviously still a bit off what you're asking.

Algae-based bioreactors are pretty recent and are a great way to produce healthy fats. Sucks that one of the best suppliers of this, Solazyme, has cut off supply to Soylent.
Hello. I'm a physicist working in agriculture, in Portugal (olive oil, malt cereal, legumes).

There are a lot of things to say about this subject, but I'll stick to the two points that were originally raised.

1. Fully automated farms are possible in theory for a small range of applications (like greenhouses). Human supervision and operation is usually necessary, mainly because of the elements and wildlife (e.g. humidity damages machines/sensors, animals chew wires/irrigation tubes, underground rocks are still invisible to surface dwelling tractors), so failure is a constant. GPS guided machines are now being widely used for sowing and pesticide application, with benefits in terms of both cost and time (fuel) reduction.

Automation in irrigation is now everywhere. One can easily find fully-automated center pivot irrigation systems.

Some fields have been adapted to allow for automation, like olive orchards. Olives can now be harvested in a fraction of the time it used to take, with a single machine and operator (a ~30 ha orchard will take 1-2 days with 1 machine and operator vs. 5-7 days for 2 or 3 machines and a team of 10; in the recent past there would be no machines and a team of 20).

There are some clear benefits in using drones that scan crops for detection of spots where irrigation fails or some fungus is starting to spread in advance, and also allow for better sampling for analysis. Besides any farmer can afford a drone nowadays.

Automation is usually expensive (for the first investment at least), and as such only large scale (corporate) farms can afford it. Cooperative family-scale farming can reduce the cost/acre (e.g. time sharing one single harvest machine) but there are pros and cons.

1.2 IoT in agriculture: please don't.

1.3 There are still fewer people in farming. I don't know if automation is the only one to blame, because either way farming is tough. Perhaps the wages may not be the most interesting for some one not already in the business, as well as the overall lifestyle. The lowest wages correspond to jobs being replaced by automation, so those salaries are dropping even more.

2. Crop yield largely depends on soil quality, proper irrigation and seed variety. Seasonal factors also play an important role, such as unpredicted disease surges. Technology already has the means to improve here, and part of what the future will be resumes to the dissemination of these already known practices.

2.1) One cannot avoid mentioning OGM seeds OGM seeds are not that impressive for me. Predicted yields from OGM seeds are said to be greater that non-OGM seeds, but the difference can be achieved alone with proper agricultural practices and cheaper non-OGM seeds. I have never tried it, but I suspect that OGM seeds yield in those ideal conditions will converge with non-OGM yields because OGM seeds are not the incredible Hulk, and non-OGM cultivars are pretty great as well. Also, pesticide resistance is not a good thing in itself, allowing for wasteful and unhealthy over-the-top usage, for example. There are many gimmicks in this area, and the possibility for patenting is the major responsible (I think).

2.2) Plasma physics Plasmas have been used to treat seeds before sowing, apparently with some good results. Still in research, I will not say it works or that it is feasible.

-

The future of agriculture can take two distinct directions from now on, in my opinion. It either moves towards scaling up the current technology (corporate agriculture is already doing this) or it will start to incorporate some rather new practices that, although not opposed to machine/chemistry based technology, just do not need them as much. The latter works for both small scale and, to some extent, big scale farming.

In the particular case of olive trees, the dynamics of its auxiliary species is well known. Maintaining the top 4 or 5 species in the local ecosystem can reduce the usage of fungicides, insecticides, as well as erosion and soil correction needs. This is not "new ...

(comment deleted)
Vertical farms are going to be huge. They're almost fully isolated from the outside environment save for taking in local air and removing carbon from it. They require magnitudes less water than traditional agriculture and no fertilizers. You can build them in the middle of a downtown core if you need to. And you could optimize power consumption so that the majority of your energy comes from renewable sources.

See Aerofarms[0] for an example. I've heard of companies in Japan developing similar technology. If I meet the right people I might take the plunge and build one myself.

Traditional farming needs to go. It requires huge amounts of land to be cleared, fertilizers, and a large amount of fresh water. Most of what goes into such farms is wasted. Then you have to ship the output over large distances. Better to go vertical in my opinion.

[0] http://aerofarms.com/

Please note that vertical farming is mostly about producing leafy greens and other low-footprint, high-value goods. Purely from the energy standpoint there is no sense in using vertical farming for the bulk of calorie consumption. Compare traditional farming:

sunshine -> green parts of the plant -> seeds -> food -> people

with vertical farming:

sunshine -> [... -> electricity -> LED light] -> green parts of the plant -> seeds -> food -> people

The part in brackets will inevitably reduce overall efficiency by at least one order of magnitude. This means you will need to cover x10+ of land area with solar panels if you want to use them for electricity for vertical agriculture.

They're mostly about producing leafy greens right now for two reasons: proof of concept and profitability. I look forward to the technology expanding to other plant types: tubers, legumes, and the like.

The efficiency of solar panels is improving but not ideal. The advantage is that the energy can be stored in batteries and the light from specialized LEDs only emit the spectrum of light that is useful to the photosynthesis process without generating excess heat. In a controlled environment you're protected from storms and pests. Using aerosolized nutrient delivery eliminates soil management. A good part of your energy profile is likely going to come from traditional sources but the cost of not having to manage acres of land and ship the final product hundreds of kilometres is a big savings.

An argument from the market: do you realize that current vertical farms should demonstrate crazy margins to justify the move to less profitable crops? According to [1], you can sell a cup of micro greens for $5. You will likely need a few times more space (and likely ten times more time) to grow a single soybean plant. The plant will produce a few soy beans that you can sell for less than a cent. How can this be profitable even in a distant future?

The theoretical maximum efficiency of solar panel is around 30%. So 70% of solar energy is already wasted compared to growing stuff in a field, which means that even if the batteries, transmission lines and LEDs are perfect (they are not), you have to cover three times as much land with solar panels as you would have covered with plants. It also seems that you would prefer growing stuff 24/7, for this one needs to compensate for the time when sun isn't shining, which means 9x land. It seems to me that vertical farming will be even more disastrous for the nature than "traditional" one.

[1]: https://www.foodcoop.com/produce

Plants aren't using 100% of the suns energy though. They use specific wavelength that the sun output. Do they use more than 30% of the suns available spectrum? I don't know.
The obvious counter to this is to not use solar panels and LEDs, and just pipe the light with pure optics--mirrors, prisms, light-pipes, etc. Split off the green light for some other use, and direct the rest onto the plants. When the sun isn't shining, feed the optics system with a large, efficient plasma lamp.

Of course, you still need to occupy surface area to collect the free solar light, but you don't need to do it over fertile soil. And you're still going to have efficiency losses, no matter what. The primary benefit of growing indoors is not in the lighting, though. It's in the other factors, like water conservation, pest management, climate control, and pollination control.

Re: LED lighting

Any idea if anyone has experimented with using heliostats and light pipes (fiber optics) to move sunlight from the outside to the inside for growing plants, while eliminating the heat issue - for growing plants?

I know it has been done for lighting a room, as well as for artistic purposes - but I am not sure about whether it has been tried for indoor farming...?

This is offset in part by reducing or eliminating transportation costs and intermediation.

I was reading an interesting article about the farmer's market in my country, production is actually not the largest cost, it's less than half of the actual sale price.

See for example:

http://capreform.eu/farmers-share-of-food-chain-value-added/

(spanish) the article I was reading: http://www.cnfr.org.uy/uploads/files/Manual_de_comercializac...

Even if it's possible to squeeze three times more money for the actual plant growing, I'm not sure it will make a dent on the difference. From the energy perspective: highest possible solar cell efficiency is around 30%, which means just this particular step in the chain will already eliminate the upside of local produce. We still have to account for batteries, LEDs, transmission lines, heating or cooling of the farm, and pumping water and nutrients around. The numbers just don't add up for "large-scale" staple crops like wheat or soy.
I see lots of interesting articles on vertical farms, but they always seem to produce only lettuce, which has zero nutritional value. it worries me that they don't seem to produce anything more substantial
Odds are many of those companies doing vertical farms are really planning on moving from lettuce to much more profitable leafy greens, like pot.