Hard Light Productions Forums
Off-Topic Discussion => General Discussion => Topic started by: Luis Dias on November 02, 2012, 04:43:30 am
-
'king amazing.
When fully autonomous systems take to the field, they will look nothing like tractors. With their enormous size and weight, today's farm machines have significant downsides: they compact the soil, reducing porosity and killing beneficial life, meaning crops don't grow so well. Compaction also increases erosion by rainwater run-off. "Why do we plough? Mainly to repair the damage that we have caused with big tractors," says Blackmore. "Up to 80 per cent of the energy going into cultivation is there to repair this damage. Surely there is an opportunity to do things in different ways." Fleets of lightweight autonomous robots have the potential to solve this problem, Blackmore believes. Replacing brute force with precision is key, he says. "A seed only needs one cubic centimetre of soil to grow - if we cultivate just that we only put tiny amounts of energy in and the plants still grow nicely." These lightweight robots could remove the need for ploughing altogether, significantly reducing the amount of energy, and thus carbon dioxide emissions, coming from farming. And with less compaction, the soil keeps its structure and beneficial organisms, and is able to absorb more water and stay fertile for longer.
Link for the article in New Scientist is paywalled.
It's here however: http://www.newscientist.com/article/mg21628882.200-farmerbots-a-new-industrial-revolution.html?
A short concept video:
http://bcove.me/ejqkqb5v
-
If they're trying to use GPS as a precision reference, they're gonna have a bad time... because it is NOT precise (ty military for watering it down for us).
Wheel encoders are also a big no-go, because they are inaccurate as hell on any terrain that isn't smooth, so they'll most likely have to figure out their positioning via optical methods or hook up some radio beacons. The video shows them working in a swarm, so they may be able to triangulate their relative position to each other and make a few rows with no trouble.
It seems to me that hexapods would be ill-suited for farm work as-is, mainly due to the fact that the leg system has a high number of joints where dirt can get in and cause some serious damage... but they do have the unique advantage over wheeled and tracked bots by potentially having the smallest afflicted surface area (regarding soil compaction). I'd like to see a version of the hexapods that employ boots to protect a few of the joints.
-Sleep post... so grammar might not be up to par.
-
GPS can be used in conjunction with wifi for relative positioning. GPS is already in a number of pieces of famr equipment (combines, sprayers) allowing them to program a digital course into the machine for the operator. It's pretty cool technology, and GPS is generally precise enough for those purposes (offset in most places is only 3-9 m).
-
Pretty sure they'll be ruggedized if they actually go into production. The ones in the video look like just proof of concept stuff.
-
Exactly, so we are talking 3 years to field testing, perhaps 10 years to actual deployment.
But think of it: all agriculture will be done this way (because theoretically it has convinced me extremely well), and people 40 years from now will talk about tractors in the same vein people now talk about asses and bulls and so on.
-
Exactly, so we are talking 3 years to field testing, perhaps 10 years to actual deployment.
But think of it: all agriculture will be done this way (because theoretically it has convinced me extremely well), and people 40 years from now will talk about tractors in the same vein people now talk about asses and bulls and so on.
How many farmers do you know? =)
Most of the ones I know are traditionalists. It's going to take one hell of a lot of convincing to entice them to move to new technology. I don't see it happening with any great speed.
-
Yeah, I mean it's not like they use any kind of machinery nowadays.
(BTW, I do have farmers in my family and while I'm 100% urban, I do help in the tasks on my vacations, just to answer your rethorical question...)
-
Yeah, I mean it's not like they use any kind of machinery nowadays.
Any idea how long it took just to get that technology into widespread use in developed countries? In principle function, the tractor hasn't actually evolved all that much since it was first built in the 1850s.
Advances are cool, but widespread adoption takes a while, especially considering capital costs of farm equipment.
-
The gap now has shrunk some orders of magnitude since the 1850s. Yes, even on agriculture. I've seen it myself, new techniques and technologies that weren't available ten years ago are now considered "standard". So I'd say we kinda disagree violently on this one.
-
Having grown up in agriculture the issue with any new tech is the same as any other industry which is return on investment which is generally measured in years for vehicles. also it dont matter if it is twice as efficient if it is three times as expensive
-
Surprise. :rolleyes:
Honestly Luis, if you see widespread rollout of this technology in even a decade, then I admire your exuberant optimism. You're talking about convincing a breed of the most stubborn, traditional people on Earth to change the very core practices of their business based on theory, without regard for capital cost.
-
Thanks Headie, that's my idea as well. It's all about the profits, and MP, who said anything about "without regard for capital cost"? Clearly, leaving work for bots instead of people will be super sweet to any farmer's profits.
We are on the brink of wide robotization of every production line, be it cars, carrots or iPhones.
-
Automation still needs to be cheaper for it to take off.
You're talking about building a fleet of robots (a fairly large fleet per operator, considering the size of agricultural plots in some parts of the world) that need to operate in harsh conditions ranging from near-zero temperatures to upwards of 40 degrees C, in rain/sleet/snow/sunshine, be impervious to high-wind blown dust, and generally be rugged enough to do farm work. They have to be cheape enough to be replaceable when a unit fails, or easily repaired on site - which will have to be by trained staff with some specialty knowledge.
Not saying it won't be possible one day, just that there are a lot of economic and feasibility hurdles that have to be solved first, and convincing farmers that this is better is going to take some time. You've had to prove that the costs of purchase, operation, maintenance (human salaries and parts), and replacement total to a lower dollar figure than traditional farming, and the chances of that being done on a widespread basis inside a decade - or even four - are slim.
Factory farms might do it first, for obvious reasons, but a lot of crops are still grown outside a factory setting by individuals or small corporations spread out over wide areas.
-
You can't tractor pull with robots...
-
Oh? I'm sure we'll find a way. :p
-
Oh? I'm sure we'll find a way. :p
there is always a way :drevil:
-
All the evidence farmers need is the big bucks going for their competitors. The hurdles are a matter of time. The costs should be considered alongside what they are substituting, which is generally a huge machinery fleet costing a lot of cash, and a lot of man-hours.
-
And you seriously see this technology being deployed on a wide scale in a decade?
-
And you seriously see this technology being deployed on a wide scale in a decade?
Each Hexabot is probbably around $1,000~2,000 each. I don't know what the average cost of farm equipment is, but if it's cheaper than getting a new tractor (both in maintenance/fuel costs and initial purchase) then you can bet on it.
-
Your endurance and speed with those robots will never work for large-scale farming, ever. An autonomous tractor might be expensive, but it would get the job done. You can also use the tractor for field maintenence, etc. Those concept robots, however, would only work for planting, maybe. I see that as a novel concept in networking and autonomous systems, but that's about it. For any large field operation, your conentional machinery, manned or unmanned, will remain the dominant working vehicle(s).
-
Ahhh now I see the equivocation.
No, I don't see "widespread" deployment in 10 years. I see "initial" deployment in ten years. Widespread? 20.
-
I see your deployment of that sort of system... as never being a viable one with resepct to cost or benefit. The concept robots act in a manner similar to that of a primitive group of farmers, each tending to each individual seed at planting and harvest (and for many crops, the farmers would scatter seeds rather than plant them regardless). The method is slow and time-inefficient, and to make up for the time, you'd need more working units.
And then the problem becomes power and propulsion. Conventional batteries tend to be fairly heavy and not terribly efficient. Exotic batteries contradict their "green" proponents, who like to ignore the wasteful, lossy, and environmentally unfriendly processes of refining rare-earth metals...
So ultimately, this may work for... I'm not sure what, but small-scale, high value crops. But the cost of the system will have an impossibly hard time paying for itself. You'd be better off hiring farm laborers and giving them a reasonable wage; with an expanding population, you'd also be more ethical in giving people work as well. My conclusion remains this: this is an interesting study in networking and programming, but the viability of the system in this form is limited at best. If you were going to use robotic farmers, you might as well jump right ahead to hydroponics, which would be an even better long-term research project.
-
I think what's really ethical is putting robots doing all of the work and leave us all playing Freespace.
-
Your endurance and speed with those robots will never work for large-scale farming, ever. An autonomous tractor might be expensive, but it would get the job done. You can also use the tractor for field maintenence, etc. Those concept robots, however, would only work for planting, maybe. I see that as a novel concept in networking and autonomous systems, but that's about it. For any large field operation, your conentional machinery, manned or unmanned, will remain the dominant working vehicle(s).
The bots where designed with one thing specifically in mind: to reduce the compaction of soil by replacing larger heavier machinery with smaller lighter machinery. The theory is that this will reduce the need of tilling and plowing the soil while at the same time reducing the amount of runoff wastewater - a major contribution to water pollution.
Fertilizer is applied at the time of sowing, or before hand. But, it can also be mixed into the water used for irrigation. There are already irrigation systems in place that pivot around a water main, and do not significantly contribute to soil compaction (vs. the harvesters and tractors). Plowing and tilling will still need to be done due to the weight of the harvesters, but it would still be reduced by removing the sowing tractor from the equation. We'll see how well this theory holds out as the agriculture universities run it through their testing fields.
Regarding power: the robots can have solar panels strapped or painted onto them, which will extend the life of the needed battery. Yes, lithium polymer batteries are highly toxic to the ecosystem. That's why we recharge them, and reuse them as much as we can.
Regarding ethics: It's up to the farm owner to decide how he or she runs their farm, and NOT the engineers. Just because the technology exists does not mean that it should be the preferred solution.
-
People tend to underestimate the amount of work autonomous processes do when they are not limited by work schedules (sleep, rest, etc.).
-
'king amazing.
When fully autonomous systems take to the field, they will look nothing like tractors. With their enormous size and weight, today's farm machines have significant downsides: they compact the soil, reducing porosity and killing beneficial life, meaning crops don't grow so well. Compaction also increases erosion by rainwater run-off. "Why do we plough? Mainly to repair the damage that we have caused with big tractors," says Blackmore. "Up to 80 per cent of the energy going into cultivation is there to repair this damage. Surely there is an opportunity to do things in different ways." Fleets of lightweight autonomous robots have the potential to solve this problem, Blackmore believes. Replacing brute force with precision is key, he says. "A seed only needs one cubic centimetre of soil to grow - if we cultivate just that we only put tiny amounts of energy in and the plants still grow nicely." These lightweight robots could remove the need for ploughing altogether, significantly reducing the amount of energy, and thus carbon dioxide emissions, coming from farming. And with less compaction, the soil keeps its structure and beneficial organisms, and is able to absorb more water and stay fertile for longer.
Link for the article in New Scientist is paywalled.
It's here however: http://www.newscientist.com/article/mg21628882.200-farmerbots-a-new-industrial-revolution.html?
A short concept video:
http://bcove.me/ejqkqb5v
Looks intriguing. I can see a single unit costing upwards of several thousand dollars if built from off-the-shelf components but the cost will drop dramatically if it's designed to be built in any numbers that would be useful. Maintenance seems like it would be an obvious concern if you have hundreds of these running around your fields. Ruggedization is probably less of a concern than you would all think, I'm willing to bet that will be a comparatively simple step in future iterations. What would worry me about it is how well it would handle mud. The concept video doesn't show much but it definitely seems like they can have a production model ready in 10 years, assuming they don't hit any weird snags.
How fast we can see deployment is something else. Honestly I find MP-Ryan's assertion that all farmers are stubborn traditionalists quite laughable, but if it turns out he's right and all farmers just hate change that would be a big snag that could easily kill the project before it finishes. Even the ones that are willing to change won't be able to change over night just by virtue of there being a lot of overhead work that needs to be done. Completely changing how a farm works (or any large system, for that matter) just isn't something that can be done quickly.
EDIT:
Quick reference: Looking at wikipedia just the switch from steam tractors to gasoline tractors took 30 years and two real tries. Not entirely analogous to what we're talking about here, but it's something to think about.
EDIT2:
these are just pictures I thought were neat when reading the wiki page on tractors: :nervous:
http://upload.wikimedia.org/wikipedia/commons/8/8e/Wheat_Planting_Rig_May_2007.jpg
http://upload.wikimedia.org/wikipedia/commons/d/d1/Modern_John_Deere_Tractor_IMG_0401.JPG
http://upload.wikimedia.org/wikipedia/commons/f/f3/Tractor_fanguejant.jpg
http://upload.wikimedia.org/wikipedia/commons/e/e6/Trackmobile.jpg
-
Maintenance seems like it would be an obvious concern if you have hundreds of these running around your fields. Ruggedization is probably less of a concern than you would all think, I'm willing to bet that will be a comparatively simple step in future iterations. What would worry me about it is how well it would handle mud.
And this is what I'm getting at with farmer's being traditionalists. All the farmers I know or have met prefer things that are easy to use, easy to maintain, reliable, and cost-efficient.
My experience with rugged-ized electronics (and I have a fair bit now with the amount of field sampling and other work I do) is that they cost twice as much, are twice as hard to maintain or repair, and last half as long as the non-rugged alternatives. And the elements (weather) are a ***** on anything with a battery. So I see some pretty hefty technical/pricing hurdles here.
-
My experience with rugged-ized electronics (and I have a fair bit now with the amount of field sampling and other work I do) is that they cost twice as much, are twice as hard to maintain or repair, and last half as long as the non-rugged alternatives. And the elements (weather) are a ***** on anything with a battery. So I see some pretty hefty technical/pricing hurdles here.
This is all obviously true and I wouldn't really expect it to be any different. But there's no reference point for this particular system other than the non-ruggedized prototypes, so any technical or price comparison would be near meaningless. If people define good requirements for unit lifetime and maintainability and stick to those requirements when they get to that stage the only thing they'll need to worry about is cost. I didn't see anything in there about cost per unit, but yes, that could be an issue. I'm no farmer and I don't work for the people that are on this project so I don't know how much alternatives cost at all, but I'm assuming they did these trade studies before they started work on the project, it would be silly not to. Freshman/sophomore undergrad silly. (unless of course they're not serious about taking this to the market, in which case this discussion is moot anyway :p)
-
why are we still growing things in dirt? why hasnt hydroponics gone mass-scale yet? those kinds of facilities would be prime targets for robotization. most of them are more or less use automated environmental systems, you just need to automate things like planting and harvesting.
-
Because dirt is (I'm sorry for this) dirt cheap? :p
-
But that would make sense Nuke!
-
i guess its an art only known by people who grow dope.
-
why are we still growing things in dirt? why hasnt hydroponics gone mass-scale yet? those kinds of facilities would be prime targets for robotization. most of them are more or less use automated environmental systems, you just need to automate things like planting and harvesting.
http://en.wikipedia.org/wiki/Hydroponics (http://en.wikipedia.org/wiki/Hydroponics)
Looks like not all plants like being in just a water solution. There's also the issue of wet-rot and diseases.