Tree Planting Robot design

[Scooter]

The wave of the future?

http://www.yankodesign.com/2010/08/19/r ... ave-earth/

Discuss. Or perhaps I should say, "criticize."

PS: Thanks Natalia ...

[Nate]

I like how they inject steam to kill off competing vegetation (that'll work) and can bag up less than a 90 pound girl.

[Scooter]

By criticize, I meant let's "debate the engineering feasibility."

The steam tank is interesting. It holds forty liters of steam, or more likely, compressed water that is stored at greater than one hundred degrees Celsius.
- What materials are used for insulation to minimize heat loss to the rest of the robot, both because it is a waste of energy, and because the unit should probably not be allowed to overheat (temperature sensitive components)?
- What is the weight load? 40 litres of water is normally 40 kg of weight, or 88 pounds. Add the weight of 320 trees (let's say around 30 pounds if they haven't been watered excessively). Plus the weight of the robot. At that size, even using lightweight composites (and the battery is always relatively heavy) the robot probably can't weigh less than 30 pounds if it is going to be durable. So the overall weight of a loaded unit must be at least 150 pounds, and maybe approaching 200+.
- Does the steam get supplied externally from the trailer that the robot is transported upon? If so, how large is the water supply tank?
- What energy source is used to create the steam? Water has a very high specific heat, much higher than many other liquids. What this means is that it takes a lot of units of energy to raise the temperature of water, in very simple layman's terminology. The battery in the robot will not be sufficient to take water from normal temperature to a sufficiently high temperature (battery drain in general is going to be another set of topics). Does the trailer unit supply water that was heated in town, or does it heat the water before being added to the robot? How does the high-pressure system maintain integrity, and minimize energy loss during transfer of hot water to the robot? If the trailer unit transports water that is already heated and compressed, does it need TDG placards?
- What is the effectiveness of the steam in killing competing vegetation? Plants are pretty durable. I don't think that a brief burst of pressurized water at a temperature just above boiling would kill a lot of competing vegetation. I suspect the water would have to be much, much hotter than 100 degrees. Also, is the volume sufficient? 40 litres for 320 trees is only 125ml per seedling. Can the water be hot enough, and ejected at a high enough velocity/pressure for 125ml to effectively eliminate competition?

Tomorrow we'll talk about the battery, and the next day, about the propulsion system (why four legs instead of six or eight?). This will give me something to think about while I'm planting tomorrow.

[Nate]

Steam for vegetation control is not at all a developed/scalable technology at this point. In addition the logistical problems of the volume of water required (as you point out, it's 125ml/seedling with that design) the water needs to be heated to somewhere between 300-400 Celsius, if not higher, to break down the vegetation appropriately. The more effective models use propane/flame or whatever to heat the steam.

Weed control is probably the number one issue facing farms/growers who want to be certified as "organic", so there's a lot of people who'd like to see steam technology become a better and more practical reality. Unfortunately it's not there yet, though there area few industrial machines kicking around that work off of a truck/wand system or as a trailer on a three point hitch.

[Mike]

How fast does it plant, and what it's cost per hour running?

A planter costs wages + other camp related costs per hour/day for some number of trees. Even if the robot works well (which I think we'd all agree is a huge if) it could prove that planters are cheaper anyways...

[Scooter]

This is only a conceptual design, as far as I know. I actually don't know the background, but I suspect it was just created as a "what if" sort of project. At this point, yes, human planters are probably massively cheaper and more effective. Although on a very flat farm field sort of environment, with GPS technology embedded to control ground coverage, a robot like this might prove to be cost effective in limited circumstances. That's just a guess on my part - I'm envisioning a scenario with relatively flat terrain, limited obstacles/slash, and no competition (hence no need for the steam bath).

[granola]

Well, they already have these:

http://blog.overkillinterstellar.com/#post9

These plant 200-600 trees per hour.

And these:

http://www.youtube.com/watch?v=wJSBxXE23Ck

[newforest]

I compete with machines all the time. Then it rains.


I got a nice planting job on an organic Ginkgo farm once. They used flame as a weed control, but this stressed the tree enough to cause losses. I'm sure paying someone as absolutely little as possible to run the flame machine couldn't have contributed to those losses. Machines also can't do a fill-in plant it turns out. The original spacing was 12 inches by 18 inches.

So I got to plant the oldest tree species on the earth for the first and likely only time. My crew walked 9 rows simultaneously and we kept flats of seedlings on a metal rack on the back of a tractor. We invented a game.....guess the next Classic Rock artist to be played on the radio, win free pizza that night. Easier than you might think, 7 hours in.

We did well at the job. It was like planting a corn field, mindlessly listening to mindless music, like machines. Where there was a hole with space for one brand new Ginkgo seedling, we planted one seedling. Where 3 were needed, we planted 3.

Unfortunately another contractor was on the job, and he bid a penny less than I did. His crew planted 2 or 3 everywhere a new seedling was needed, and occasionally simply took handfuls of seedlings and randomly threw them a random direction in the field. We watched. The other contractor was cheaper by the tree planted of course and even though I was cheaper on a per acre basis doing the job correctly, Corporate directed that the other contractor be hired for the next year, as that would be cheaper to plant that year's contracted seedlings, though it would of course finish less acres. Their field supervisor told me he wanted to quit his job, but couldn't. If only Corporate could be replaced by non-thinking robots. Oh, wait.

[newforest]

oh and I had to read about the tree planting robot:

"Maintenance costs would be cheaper than benefits, overtime, breaks, health insurance, HR, 401k's, and other human support costs."

that part made me laugh.

[fluffer]

The bucket attachment that sticks a tree into a mound is a good idea. I've always thought mounding (especially the huge ones) was a bit wasteful. I mean, a good operator might do, what, 180 mounds per hour, at best? What is their rate? 150$/hr?

Otherwise it seems like most blocks would almost require space age technology for a machine to have enough sensors to pick the best places to plant while moving at a good speed. If super brushey then the bigger the machine, the more nimble and agile it must be, the more complicated it becomes. Not to mention operating and maintenance costs on that type of technology. That is for some precision planting with little disturbance, similar to what planters do. Otherwise you can have larger machines that are much slower than planters that clear the ground a little bit. Other bigass machines that destroy and clear everything in their paths and plant fast.

Clean prepped mostly homogenous ground is where I can conceive a machine potentially matching a planter in overall cost and effectiveness.

[RPF]

Never say never. Remember not so long ago some folks swore that machines would never replace horses for moving people around, or helping farmers till & harvest their fields...
Besides, with the low number of new workers choosing planting as a long term career, perhaps a machine like this is not too far off in the future.

[E.E]

If in the future -I- could become a tree planting robot via cybernetic joint enhancements and muscle grafts and such, that would be pretty cool.

[bdbrown]

Uhm, never!.. well, maybe a machine could replace some peoples trees, but there is a level of skill and production that will never be replaced by a machine.. never

[Nate]

Never say never. Remember not so long ago some folks swore that machines would never replace horses for moving people around, or helping farmers till & harvest their fields...
Besides, with the low number of new workers choosing planting as a long term career, perhaps a machine like this is not too far off in the future.

Maybe some sort of machine, but not this goofy contraption.

[salbrecher]

Direct seeding of blocks at the time of trenching/stumping/screefing is currently being studied in BC and elsewhere and shows positive results. It costs roughly 18 cents total for direct seeding via machine vs 38 cents for the planted tree. Machines planting seedlings will likely never replace planters in BC on steep/inoperable ground. Direct seeding on blocks where machines can operate may however become increasingly utilized.

http://www.forrex.org/sites/default/fil ... art21.pdf4

http://www.phd-forestry.se/Pekka-Helenius.pdf

[chronic]

However the flip side to that, having done tons of surveys is that there can be sporadic densities throughout blocks with NSR areas that have to be planted later on anyways, often in tough to reach places driving up costs because of access issues, walk in areas, ect... A lot of times direct seed blocks have very high densities and will require thinning and spacing anyways, so while companies may save some money on planting, there is a strong chance of other costs coming into play in the future.

[krahn]

If in the future -I- could become a tree planting robot via cybernetic joint enhancements and muscle grafts and such, that would be pretty cool.

this is not just sci-fi i'm sure we'll be using some sort of mechanical assistance one day. at first they'll be a bit bulky, and only used by those companies that require it for safety. but then one they'll double our production. maybe. hopefully i'm around to see it, and make a comeback.

[krahn]

i can't see machines matching the cost efficiency a good vet can get, if you count all the various factors, including a lack of healthy care for the human. that is, i can't see that happening any time soon, with seedlings, not for quality trees on difficult land. if they start using seed more, that's another matter.

i'm sure MIT could build something now to compete with us, but it would cost millions in research and millions more to construct.

[mwainwright]

yeah this particular design seems a bit goofy at the moment, but i'd put my money on drone-delivered trees in the not too distant future.

[Pandion]

http://www.cbc.ca/m/news/canada/british ... -1.3205159

Not much to worry about yet.

[jdtesluk]

Wow. What a couple of clowns. I wonder if this was a tongue-in-cheek project. 15 minutes for ten trees in a straight line? It looks like a glorified shopping cart, with some mods done by the A-team in the last 5 minutes of the episode. Fun maybe for Robot Wars but I have serious trouble believing anybody with an ounce of forestry pedigree could have been impressed by this design.

The insulting part of this crap is that they offer this as a way of assisting workers. On that basis, their project should get an "F". Such a machine, even fully refined, would simply remove the easiest ground from conventional planting, reducing the amount of biddable work for human crews, and leaving them with only the most difficult shnarb. On top of that, they use a dibble techniqe of driving a spike into the ground to create the hole before dropping the seedling, which means the plug goes into compressed earth that is harder for the roots to penetrate. Couple the "F" for the social utility with an "F' for silvicultural utility.

Won't be seeing my Indiegogo dollars.