"With each detonation, [it] loses just one or two legs."

November 20, 2012 6:45 AM

A simple, beautiful solution to clearing landmines in Afghanistan. From the public filmmaker competition section of Focus Forward, a series of documentaries about people who are changing the world.
posted by bwerdmuller (80 comments total)

The only thing is, with such non-systematic mine-searching you could never be sure that you didn't miss quite a bunch of mines. Nevertheless, a beautiful and innovative approach!
posted by SAnderka at 6:54 AM on November 20, 2012 [10 favorites]

It would be interesting to see if there were a way to make it more systematic.
posted by bardophile at 6:56 AM on November 20, 2012 [3 favorites]

My brain was trying to link this post with the one below it.

(Since this computer is running IE8 they don't want me to read their web site.)
posted by Jahaza at 6:59 AM on November 20, 2012 [2 favorites]

I can't be the only one who is thinking of 'Katamari Damacy' here...
posted by qcubed at 7:02 AM on November 20, 2012 [5 favorites]

I had the same concern as SAnderka. Personally I wouldn't feel comfortable going into an area where mines had been cleared by one of these. Not only is it unsystematic, but I wonder whether it would have a 100% detonation rate across all models of antipersonnel mines, including those that had been sitting in the sand for 30-40 years.

Landmines are an abomination. They should be banned. In fact, they are banned, with the US (of fucking course) being one of a few countries that hasn't agreed to forgo their use. Using landmines should be considered equivalent to using chemical or biological weapons.
posted by Scientist at 7:02 AM on November 20, 2012 [25 favorites]

It would be interesting to see if there were a way to make it more systematic.

Ideal would be if no one put mines in the ground in the first place, but given that there are vast regions seeded with mines - and no one really all that concerned with systematically removing them - any mine that one of these things blows up is one less mine for a kid to step on. That it's far from perfect should not be a reason to not give it a try anyway. it's a nice, clever engineering solution to boot.

The documentary, on the other hand, as a piece of filmmaking was completely unimaginative.
posted by three blind mice at 7:02 AM on November 20, 2012 [5 favorites]

The Invisible Bicycle Helmet - oh wow. they actually did it. I can't wait to see one!
posted by the man of twists and turns at 7:04 AM on November 20, 2012 [8 favorites]

I'm glad this solution does not involve baby mops.
posted by jimmythefish at 7:05 AM on November 20, 2012 [11 favorites]

In the NYT from 1996: "The English have 11 million mad cows and Cambodia has roughly the same number of equally mad land mines," The Cambodia Daily said. "Surely the solution to Cambodia's mine problem is here before our very eyes in black and white."

The paper added, "The plan is simple, practical, and will make mincemeat of the problem overnight."
posted by Daddy-O at 7:05 AM on November 20, 2012 [28 favorites]

The documentary also failed to actually show this thing detonating even one mine. I mean it showed an explosion happening underneath one, but that was clearly staged for dramatic effect. Not that we don't need a cheap way to get rid of mines, and hooray if this is it, but I remain unconvinced.

Has there ever been an X-Prize-like contest for developing cheap, effective, safe mine-clearing technology?
posted by Scientist at 7:07 AM on November 20, 2012 [3 favorites]

"Landmines are an abomination. They should be banned. In fact, they are banned, with the US (of fucking course) being one of a few countries that hasn't agreed to forgo their use. Using landmines should be considered equivalent to using chemical or biological weapons."

I agree they are an abomination, and that it's a good thing they're banned.

However, I see one current reason why both the US and South Korea are not refusing to use them--that of the DMZ. Seeing as my extended family is relatively close to that border? I'm willing to make a deal with the devil here.

If reunification ever *does* happen, hopefully the DMZ will have been the last, final place for mines to have been deployed en masse.
posted by qcubed at 7:07 AM on November 20, 2012 [1 favorite]

In the NYT from 1996

That observation may have been the inspiration for a board game.
posted by jedicus at 7:08 AM on November 20, 2012 [1 favorite]

I like how in the Invisible Bike Helmet video it's 2 minutes and 50 seconds of "inspiring" closeup pans and open design spaces and then 5 seconds of showing the actual product (which looks even more uncomfortable than wearing a helmet).
posted by DU at 7:10 AM on November 20, 2012 [4 favorites]

It should dispense bread crumbs as it goes along, so there would be a path to follow.
posted by StickyCarpet at 7:11 AM on November 20, 2012 [9 favorites]

qcubed: I thought that the threat to South Korea was not so much the North Korean army, but rather the massive artillery batteries constantly pointed at Seoul, like a gun to a hostage's head? I don't see how removing those mines would create any kind of serious threat to South Korea, and I certainly don't see why we should plant new ones. I am open to attempts to persuade me, as I freely admit that I am far from an expert on the subject of North-South Korean relations.

Yeah also that Invisible Bike Helmet documentary is dumb and the product itself is pretty underwhelming. "We're going to save the world," indeed.
posted by Scientist at 7:15 AM on November 20, 2012 [3 favorites]

I see a lot of familiar names on their site--I've worked with quite a few of these people. Most of them are not "how-to" doc makers, meaning they din't direct segments you'd see on the DIY channel, which might actually be better for explaining the tech or engineering behind some of these innovations. But it's hard to get big time filmmakers to want to crank out a short that just explains something,thus the artsy bits.
posted by Ideefixe at 7:19 AM on November 20, 2012

Sorry for the serial commenting, but I am also not sure I trust the general design of this mine-clearing katamari thing. A lot of mines are quite small (e.g. buttefly mines and the related dragontooth mines) and seem to me like they would be likely to just slip between the pads of this thing as it rolled along, without being detonated.
posted by Scientist at 7:22 AM on November 20, 2012

A simple, beautiful solution to clearing landmines?

Take all the folks who authorize the deployment of landmines, the folks who authorize the procurement of the landmines, and the board members and executive officers of the companies who manufacture the landmines, and frogmarch them across the minefields.
posted by snottydick at 7:31 AM on November 20, 2012 [27 favorites]

If you strung 50 of these things together on an axle of some sort, and powered it somehow from far enough behind that the drive system would not be damaged by the explosions, you might have something more systematic. You could probably make it a little heavier too.
posted by COD at 7:34 AM on November 20, 2012 [1 favorite]

I love how these things are "powered by the wind"... so basically they will all get blown into a ditch on one side of the minefield. Great.

The power of a good videographer, that it can make such a dumb idea look plausible.
posted by anthill at 7:37 AM on November 20, 2012 [2 favorites]

It would be interesting to see if there were a way to make it more systematic.

At the very least, fit them with a GPS so there's a plotted, archived track. I agree this is kind of useless but it's a start.
posted by jimmythefish at 7:38 AM on November 20, 2012 [6 favorites]

This is my absolute favorite of the series, it's a huge deal if this works as they claim. Using scorpion venom to carry light-emitting substances to brain tumors so a surgeon can actually tell tumor tissue from healthy tissue with the naked eye during surgery, rather than flying pretty much blind and relying on pre-surgery scans like they have to do now. HUGE deal.
posted by jason_steakums at 7:41 AM on November 20, 2012 [3 favorites]

Yeah also that Invisible Bike Helmet documentary is dumb and the product itself is pretty underwhelming. "We're going to save the world," indeed.

But they're "girls"!!!
posted by The 10th Regiment of Foot at 7:47 AM on November 20, 2012

Yes but a sphere's contact area with the ground is quite small, you'd have to make numerous multiple passes and have a way to have it pass over ground just a few inches offset from the last past. Something like a long chain of these things being driven around a central point or something slowly taking in the chain. Think like the old self-propelled, self-mowing lawn trick.
posted by laconic skeuomorph at 7:48 AM on November 20, 2012 [2 favorites]

I wonder if there's a mine-discovery solution that would be able to be remotely controlled. Just set up the control of the remotes as a game, with high scores for clearing areas, and use first-world boredom to solve the problem.
posted by dubold at 7:49 AM on November 20, 2012 [17 favorites]

I agree this is kind of useless but it's a start.

Why useless? It won't clear minefields by itself, but every mine that blows up under one of these things is one that doesn't blow up under one of the people doing a systematic clearing later. If my job is being one of those people, reducing that probability seems useful to me. Add GPS and you at least know where it hasn't been, and with a bit more experience you can get ideas of false-neg rates and such.
posted by Vetinari at 7:49 AM on November 20, 2012

Not being able to watch the video on the machine I'm using, I went looking for this thing in a Google image search. Boy, was that heartbreaking. Landmines are pure, incarnate evil.
posted by Phyllis Harmonic at 7:52 AM on November 20, 2012

Flail tank, an idea dating back to at least WW2.
posted by Daddy-O at 7:53 AM on November 20, 2012 [4 favorites]

Yes, adding a bunch of electronics to the land-mine clearing thing will make it more effective on a per-unit basis.

No, adding a bunch of electronics to the land-mind clearing thing will NOT make it more effective on a per-dollar basis. Plus it can't be assembled and repaired by kids in the village.

Since those are the design goals, you now have an answer to the question about why it doesn't have a bunch of electronics onboard and can hopefully find peace with that.
posted by DU at 8:16 AM on November 20, 2012

Some mili-irony here: minefields are not usually employed to kill soldiers. They are laid out in an obscure pattern and concealed, but the area where they are deployed has to be clearly marked for the minefield to be useful: it's function is to deny the use of a certain area to the enemy, or to channel his movements elsewhere.

The enemy then uses various techniques to make a path, or they go around the minefield, which of course slows them and channels them. Soldiers are trained for this, and there are surprisigly few casualties in this sort of endeavor.

So, minefields don't usually kill many soldiers, but they do kill civilians--after the battle, after the war, after the living memory of their deployment has faded into bitter stories told by old people.


More irony. This was a haunting and lovely film with an attractive subtext of low-tech empowerment, responding to generations of violence. The device is clever and yet depressingly useless because of its inefficiancy.

I offer prospective followup film, with an image of a mother or father carrying a legless child to an aid station, and a dialogue strip at the bottom of the screen that says, "But we ran the device over the area several times."
posted by mule98J at 8:17 AM on November 20, 2012 [1 favorite]

Why useless?

If your end goal is returning a field to use for a farmer or a school or a homeowner, your engineer needs answer these questions:

How long will it take?
How much will it cost?
How risky to the deminers is it?

And, most importantly,
How certain are you that all the mines are gone?

The only question this device likely solves is the risk to the deminers, and it does little to solve the last, most important problem.
posted by bonehead at 8:19 AM on November 20, 2012

Actually, I think standard statistical sampling techniques could probably answer the "how certain are you the mines are gone" question. What proportion of the area has been traversed? How often were mines encountered? Etc.
posted by DU at 8:20 AM on November 20, 2012 [2 favorites]

How many sigmas would you be comfortable with before you went out in that field, DU? Before you sent your child and her friends?
posted by bonehead at 8:25 AM on November 20, 2012 [4 favorites]

This thread reminded me of a couple year old article from the Washington Post about empathy with robots:

Excerpt:

The most effective way to find and destroy a land mine is to step on it.

This has bad results, of course, if you're a human. But not so much if you're a robot and have as many legs as a centipede sticking out from your body. That's why Mark Tilden, a robotics physicist at the Los Alamos National Laboratory, built something like that. At the Yuma Test Grounds in Arizona, the autonomous robot, 5 feet long and modeled on a stick-insect, strutted out for a live-fire test and worked beautifully, he says. Every time it found a mine, blew it up and lost a limb, it picked itself up and readjusted to move forward on its remaining legs, continuing to clear a path through the minefield.

Finally it was down to one leg. Still, it pulled itself forward. Tilden was ecstatic. The machine was working splendidly.

The human in command of the exercise, however -- an Army colonel -- blew a fuse.

The colonel ordered the test stopped.

Why? asked Tilden. What's wrong?

The colonel just could not stand the pathos of watching the burned, scarred and crippled machine drag itself forward on its last leg.

This test, he charged, was inhumane.

Full Article
posted by Debaser626 at 8:26 AM on November 20, 2012 [16 favorites]

How many sigmas would you be comfortable with before you went out in that field, DU? Before you sent your child and her friends?

First of all, even a steamroller driving over every inch of ground out there is still going to have some chance of failing to detonate an existing mine.

But more to the point, the goal of the device is not to find every mine. The goal is to maximize the mines found/dollar ratio.

You can really tell the engineers who've been coddled by cheap access to trillion-dollar infrastructure in this thread.
posted by DU at 8:29 AM on November 20, 2012 [1 favorite]

On the Hovding about page there is the following blurb:

The law that had just been introduced in Sweden making bicycle helmets compulsory for children up to the age of 15 had triggered a heated debate on whether it should be extended to include adult cyclists too. To people like us, who wouldn't be seen dead in a polystyrene helmet, the thought that we might be forced to wear one by law was cause for concern.

At first I was aghast that anyone would publicly admit to such a degree of vanity. But then I looked up the stats for bicycle helmet use and in the US it's 20 to 25 percent according to a 2008 NHTSA report, so I suppose they are just expressing a viewpoint held by the silent majority.
posted by Rhomboid at 8:30 AM on November 20, 2012 [1 favorite]

I think COD has it right. String 50 of them together on a long line so they revolve on an axis, and roll it across a defined area, with a human crew on either end. Rinse and repeat.
posted by jfwlucy at 8:30 AM on November 20, 2012 [4 favorites]

There are two distinct facets to removing mines - mine clearance, which is basically creating a mine free path from a to b and demining, which is removing mines from an area, where the aim is the whole area is mine free and can be used again.

There are much, much better mine clearing machines. Because what they're tackling are typically anti-personnel mines, where the optimum effect is injury and not death (because it forces a rescue, takes up high levels of resource and destroys morale), it is not that hard to armour them. It is not impossible to armour them against anti-tank mines. But for lots of reasons - like the ground not being flat, or the presence of rocks - they don't have near 100% clearance rates. They are used before or after other mine clearing processes. As a military process, mine clearance is a balance between speed and safety. It is not expected to be 100% safe either during or after the process.

It is incredibly important in demining to make sure that you remove everything in the area. A process that doesn't do that is basically useless because sooner or later someone will step on the few mines that remain. A object with an uncontrolled, untracked path that may or may not blow up anti-personnel mines leaves you no better off. You still can't use the area.

What this technique does well on is autonomy and cost. Both of which are good when we're talking about demining third world countries with lots and lots of mines. But any system that involves sacrifice must do so without compromising coverage. I.e. if something gets blown off, it can't then drop the efficacy of the demining.

There probably is a technique that allows for multiple passes. But the international benchmark for demining isn't 80% or 90%. It is virtually 100%. So any technique would have to guarantee similar rates of success.

In summary: there are only time and labour intensive approaches to demining currently, both in terms of detection and removal/detonation. Most current techniques look at lowering the labour cost and/or time. There are possible routes for automation of both the detection and the removal. Commoditised robotic solutions are, I would guess, far more likely than focusing on cheap or replaceable materials.
posted by MuffinMan at 8:33 AM on November 20, 2012 [3 favorites]

Scientist:
"I thought that the threat to South Korea was not so much the North Korean army, but rather the massive artillery batteries constantly pointed at Seoul, like a gun to a hostage's head?"
That is a threat, not the threat. North Korea has a lot of those pointed at the South, and the South has many more pointed at the North. The real "threat" in this case is simply North Korea, as the stated goal of the government there has always been to achieve reunification with the South by any means necessary.

Yes, the massive artillery batteries are a threat. So is the million-man strong army (albeit somewhat starved of late). So are the tunnels under the DMZ. So are their antiquated tanks. So are the rumblings of NBC weaponry that the North has been trying to develop.

The South has essentially made their country in many aspects a fortress facing north--and mines are part of those defensive fortifications.

Just because the North could try and level Seoul doesn't mean the South would capitulate--and for the North to try occupying the South, well, they'd have to go through the minefields now, wouldn't they?

I personally don't see any reason for any other country to use land mines. In the case of South Korea's defense, however, it seems to me that a) they're already there, b) it's not worth it to remove them at this time, so I get why both the US and SKorea aren't signing that ban.
posted by qcubed at 8:41 AM on November 20, 2012

Not only is it unsystematic, but I wonder whether it would have a 100% detonation rate across all models of antipersonnel mines, including those that had been sitting in the sand for 30-40 years.

True. However, I see these being used in combination with traditional methods, as a system. A group of these things could potentially clear a large number, if not the majority, of mines in a field. A second pass would clear more. Afterwards, a traditional team could come in and clear the remaining mines.
posted by Thorzdad at 8:41 AM on November 20, 2012

How many sigmas would you be comfortable with before you went out in that field, DU? Before you sent your child and her friends?

Maybe you should clear it enough to send your goats out and later let the kids play there.
posted by The 10th Regiment of Foot at 8:52 AM on November 20, 2012 [2 favorites]

Metafilter: Come into thread to post a two-column-inch article that ran in the New York Times in 1996 about shipping mad cows to the minefields of Cambodia, because surely, nobody's seen that. Leave thread having learned that other people have seen it, and that there's a related board game.
posted by evidenceofabsence at 8:55 AM on November 20, 2012 [3 favorites]

I wonder if there's a mine-discovery solution that would be able to be remotely controlled. Just set up the control of the remotes as a game, with high scores for clearing areas, and use first-world boredom to solve the problem.

Wait, I thought that's what I was already doing with Minesweeper. That was all for NOTHING???
posted by orme at 8:58 AM on November 20, 2012 [9 favorites]

More than you ever wanted to know about demining efforts in Afghanistan:

The Landmine Cluster & Munition Monitor's 2011 report on Afghanistan (especially "Casualties and Victim Assistance" and "Mine Action")

The Electronic Mine Information Network re: Afghanistan (describing UN efforts and the 10-year extension of the 2003 plan to rid Afghanistan of landmines by 2013)

United Nations Mine Action Service - Five Pillars of Mine Action (wiki)

ESRI article on landmines and geodata - check out the image captioned "The Afghan demining database is the largest of all such installations in the world."

The International Campaign to Ban Landmines - Afghanistan: Landmine Fact Sheet (learn about the Italian Valmara 69 landmine, the Soviet butterfly mine, and dozens more -- "mainly from the ex-USSR, but also from Belgium, Italy, US and the UK.")

Mine Action Information Center -- fascinating database of "Mine Action Lessons Learned"

Map of contaminated square acres by province (not sure of the source -- the current figure is closer to 600 sqkm than 350 sqkm)

I haven't exactly digested all of that, but my first thought is that we're talking about 600sqkm in a country of 640,000sqkm -- it's not a lot of area. (Of course they keep finding new mine fields too, so that's not complete.) Unfortunately it happens to be area selected for being useful to walk on. Many of the hundreds of casualties a year are like the people in the film -- boys. I can see how if you were a local parent, you'd want to make the dangerous areas a little less dangerous while you waited for the professionals to arrive. (I wonder what the options are for just putting a fence around those 600 sqkm?)

Unsettling and macabre sidenote: the first link attributes the uptick in landmine casualties in 2010 to ... IEDs. As explosive devices designed to trigger with weights of 10-100kg, they fit the definition and get included in the statistics. So there are non-governmental organizations installing landmines at the same time there are NGOs dismantling them.
posted by jhc at 9:01 AM on November 20, 2012 [1 favorite]

I'd confidently walk through any field that's been cleared after the Hero Rats have identified the mine locations.

(would link to Anti-Persoonsmijnen Ontmijnende Product Ontwikkeling's website instead of the wiki page, but it seems down at the moment)
posted by radwolf76 at 9:03 AM on November 20, 2012 [6 favorites]

I saw this device in the Design Museum in Helsinki this past April---it's really huge and weighs about as much as a bicycle.

And qcubed, you'll be happy to know that while I was pushing it around the display room, I was singing 'la la, la la la la la la la, katamari damacy'.
posted by yellowcandy at 9:07 AM on November 20, 2012 [3 favorites]

"I thought that the threat to South Korea was not so much the North Korean army, but rather the massive artillery batteries constantly pointed at Seoul, like a gun to a hostage's head? I don't see how removing those mines would create any kind of serious threat to South Korea, and I certainly don't see why we should plant new ones. I am open to attempts to persuade me, as I freely admit that I am far from an expert on the subject of North-South Korean relations."

There are a lot of really fucked up things the US does, but this isn't really among them.

From a military perspective the landmines are pretty much all theater, as the North Koreans would use fast and ineffective ways to clear paths through the mines and not care about casualties, but it is very important theater. A real conflict exploding would involve the pretty much instant destruction of most everything of value in the peninsula with millions of deaths and trillions of dollars of damage within days to weeks. No one would win an invasion of either North or South Korea, that is part of the point. However, there is a lot more to the DMZ, and the landmines than just its theatrical role in preventing invasion. The DMZ provides an important playground for North Korean generals to play soldier in that is relatively safe for them and relatively sectioned off from vulnerable Korean civilians. In this sense the landmines help to make the playground feel real, while limiting the amount of serious bullshit North Korean generals can actually get away with, helping to both bring the conflict further from the brink while also at the same time keeping the current levels of persistent sub-conflict the North Koreans must maintain for their own senses of legitimacy contained. It also conveniently serves as something both sides can come together on condemning.

The exceptions for the DMZ, that are the only thing American diplomats need to sign the Ottawa Treaty, make an incredible amount of sense when you actually understand the role they play in promoting peace in the Korean peninsula. However, it is awfully easy for nations with no reason to care, because the US does for them, to take the faux moral stand and use it to score cheap anti-american points. This is, again, not to say that there aren't good reasons to be suspicious of, frustrated with, contemptuous of, or generally hateful towards US foreign policy; but this isn't really one of them.
posted by Blasdelb at 9:08 AM on November 20, 2012 [5 favorites]

What ever happened to frog marching our prisoners in a line through the territory to be cleared? It worked for Stalin.
posted by Renoroc at 9:14 AM on November 20, 2012

What Blasdelb said re: DMZ.

There's also the simple fact of the matter that removing them from the DMZ would require significant manpower to be diverted to a very risky zone--and not just over the mines.

People are a little twitchy there. Any incursion is viewed with suspicion, and it can quickly spiral out of control. One need look no further than the Poplar Tree Incident of 1976.
posted by qcubed at 9:15 AM on November 20, 2012

Also, while landmines as a threat to civilian populations pretty much lasts forever untill expensive demining efforts happen, landmines as a meaningful deterrent very much does not. If we stop dropping new ones in the DMZ and leave the North Koreans to be the only ones replenishing the supply then they'll do it in patterns they know, rendering the landmine effect useless. They are also not a threat to pretty much anybody but the most aggressive of those soldiers with ambitions towards being North Korea's future leaders, whose herd we probably want to thin anyway.
posted by Blasdelb at 9:21 AM on November 20, 2012 [1 favorite]

Keep in mind that all mines present a hazard to the EOD people who have to clear them. If running this thing haphazardly around in the mine field detonates a mine, that's one mine less to locate and clear.
posted by Harald74 at 9:48 AM on November 20, 2012

I have spent all of two minutes thinking about this problem, but I am happy to announce my sure-fire center-pivot irrigation-based landmine detection and removal venture is now accepting capital from any and all comers.
posted by Alvy Ampersand at 9:49 AM on November 20, 2012

There is an additional problem with old mines in that they're not reliable. Someone stepping on one today might not set it off, but someone stepping on one next week — boom. Same with various types of UXO, e.g. cluster munitions.

Still, the device is probably better than nothing, and that might be the alternative.

Re land mines in general, they are "area denial" weapons. (In current US-centric military jargon this is referred to "tactical Anti-Access/Area-Denial" or "tactical A2/AD".) Their purpose is to keep an enemy out of a particular area, or at least make that area less attractive than other areas or routes. There are other types of area denial weapons, but they are not much friendlier: lingering nerve gas, certain types of biological agents, and some tactical nuclear weapons are all also area denial weapons. Given the alternatives, it's not hard to see why mines are the beauty queen of that particular leper colony.

The situation on the Korean peninsula is much like Germany in the 50s and 60s, with the North Koreans playing the part of the Soviets. They have a comparatively large standing army, and a stated interest in mounting an invasion; the South Koreans and US forces would prefer to counter this threat with as little ongoing manpower as possible. For years, the NATO posture in Europe did this through a heavy reliance on chemical and tactical nuclear weapons (e.g. nuclear artillery and later SRBMs pointed at the Fulda Gap and other probable invasion routes, as well as nuclear land mines and chemical agents). A relatively small defensive force was expected to use these weapons to fight a delaying action, buying time for a full NATO mobilization and response.

Without the benefit of some of the Cold War's nastiest engineering marvels, the case for land mines — or at least an understanding of US/Korean discomfort at giving them up — in South Korea becomes more clear. Without a plausible way of delaying a North Korean invasion long enough to call up reserves (or bring in forces from other countries), it might appear that a massed attack could be a fait accompli. (Whether it really would affect an invasion is less important than the perception, on both sides, that it would change the calculus.) A change in balance is not something that anyone wants.

So until there is a functional replacement for land mines in that particular scenario, I'd expect them to stick around.
posted by Kadin2048 at 9:52 AM on November 20, 2012

The Hacker News thread on this topic is actually very informative, especially the contributions by _djo_.
posted by Spanner Nic at 10:13 AM on November 20, 2012

bonehead: How many sigmas would you be comfortable with before you went out in that field, DU? Before you sent your child and her friends?

Given that those same farmers are using that land anyway? I'd say 1 sigma would be a significant improvement.

People are posting in this thread as though the civilians in landmined countries have the option of just avoiding the landmined areas - like there's some sort of police tape strung around all the mine fields in the world.

It ain't like that, folks. If there's 10,000 land mines left in your 50,000-square-km country, there's one every 5 square km on average. Somewhere. You need to plant a field. Here's a plow and some oxen. Good luck.

Me, I'm going to wait for this thing to roll around a bit, and maybe send my oxen down a down-wind line I've posted, catch it when it comes to the line, and roll it back up to the oxen-tested upwind line. Won't be perfect, but beats the hell out of your method of not using these unless they are 6-sigma perfect.
posted by IAmBroom at 10:18 AM on November 20, 2012 [4 favorites]

DU: You can really tell the engineers who've been coddled by cheap access to trillion-dollar infrastructure in this thread.

Probably not engineers. We tend to be sensitive to cost-benefit analysis - and statistical success.
posted by IAmBroom at 10:21 AM on November 20, 2012 [2 favorites]

If this minesweeper is effective, expect insurgents to begin using it to clear a path for operations. Then expect US developers to develop a patch for landmines' 20-cent microprocessors to detect the difference between a sweeper and an insurgent.
posted by surplus at 10:55 AM on November 20, 2012 [1 favorite]

"With each detonation, [it] loses just one or two legs."

aw man, before looking at the link, I was hoping for a gigantic mecha-millipede.
posted by juv3nal at 11:03 AM on November 20, 2012 [1 favorite]

"This refugee's imperfect but ambitious prototype is nowhere near as effective or well-designed as the mine-removal/detonation idea I don't have and won't build."
posted by Uther Bentrazor at 11:04 AM on November 20, 2012 [9 favorites]

I have a picture of this on my phone. Didn't know it was the only survivor.
posted by Chuffy at 11:12 AM on November 20, 2012

Those of you who don't like this project are crazy people.
posted by TheTingTangTong at 11:21 AM on November 20, 2012

If there's 10,000 land mines left in your 50,000-square-km country, there's one every 5 square km on average. Somewhere.

Hmm, it doesn't sound like that's true in Afghanistan from the articles I linked above. "The Mine Action Coordination Centre for Afghanistan (MACCA) estimated the number of remaining hazards as of 31 March 2011 at 6,545 [sites] covering 627km2 and affecting 2,056 communities." In 2012 they're down to 563 km2 -- that report goes into a ton of detail about what they're doing and how they're getting there.

Those estimates are obviously incomplete -- there are minefields they don't know about. But casualties per year are less than 20% of what they were 10 years ago, so they must have a pretty reasonable fix on the problem. When we're talking about Afghanistan having a substantial ongoing problem with landmines, we're talking about on the order of 0.1% of the surface area, not 100%.

On the other hand, that same "2012" link says that battlefield and village-by-village searches (mostly surface-only) are recovering just an unbelievable amount of unexploded ordnance and small arms ammunition outside of minefields -- and those non-mine explosives are causing almost 90% of the remaining casualties. So the real question might be whether this is an effort-effective way of finding unexploded ordnance and ammunition.
posted by jhc at 11:31 AM on November 20, 2012

Plus it can't be assembled and repaired by kids in the village.

Since those are the design goals, you now have an answer to the question about why it doesn't have a bunch of electronics onboard and can hopefully find peace with that.

Why do you assume people in villages aren't capable of repairing electronics? We're basically talking about phone parts for a GPS tracker. Though I think you could probably do it cheaper with as much granularity (or more) by using a video camera on a tethered weather balloon and plotting that.
posted by oneirodynia at 11:39 AM on November 20, 2012

My armchair critic's criticism about this invention would be the narrow area it would sweep out and the fact that it doesn't mark out the area it has swept.

Someone above mentioned flail tanks. There is a Wikipedia page on them, they are still used today and if you're interested in them, check out the article right here.
posted by Mokusatsu at 1:11 PM on November 20, 2012 [1 favorite]

Won't be perfect, but beats the hell out of your method of not using these unless they are 6-sigma perfect.

The problem is that you can't just treat this as a typical statistical inference problem. It's going to be very hard to convince someone to pay for/accept a technique which removes most of the mines most of the time. The reaction you will get when you tell someone that there's only a 1 in 10,000 chance of losing a life to an undetected mine is that that's one chance too many. People are very reluctant to do cost-benefit when it comes to safety, in my direct experience. They want assurances, not inferences.

Mechanical methods can work, but are not good enough to work alone. From the Mine Action Information Center at JMU:

Why not use machines to make demining faster and safer?

Steel wheels capable of absorbing the effects of anti-tank mines enable this machine to cut vegetation safely.

Numerous mechanical systems, ranging from flail systems to soil grinders, are operational worldwide. However, no mechanical system has reached 100-percent clearance reliability due to the complex nature of the mine/UXO (unexploded ordinance) threat and variable terrain conditions. The term "mechanical assistance" explains that these systems, while complementary to manual deminers and mine detection dog teams (MDDTs), cannot yet replace them in their hazardous tasks. By using mechanical means to clear vegetation and process the ground, the perimeter of the mined area can be ascertained quickly, thereby making it possible for the MDDTs and manual deminers to focus on locating and destroying individual mines and UXO in a safer, quicker and more economical manner.

Mechanical flails are typically 60% to 80% effective, mostly depending on terrain and vegetation. The tanks and heavy vehicles they are mounted on are often quite expensive, and only available to governments. The Aardvark is one of the more economical ones currently being used in Afghanistan.

That all to say that statistics are not so much your friend when it comes to landmine clearance and mechanical techniques only go so far. Boots on the ground is still the preferred verification method.

Further, as the designer points out, his windblown dandelions would mostly be useful as a cheap way for residents to map out possible fields. That use is probably pretty reasonable, in fairly limited conditions (vegetation?, hills?). However, turning this into something that it's not, a primary demining tool---let alone a good-enough whole solution as some above seem to be suggesting---is, in my view, a flight of fancy.
posted by bonehead at 1:12 PM on November 20, 2012 [1 favorite]

Lots of excellent links here, no time to review them all, so I'll ask directly:

Considering the amazing positional tracking of quadrocopters, can one be equipped with some sort of finely tuned low energy ground-penetrating radar? It could map out an entire minefield automatically from a very low altitude and transfer the GPS data to a ground-based robotic flailer. Differentiating between a mine and a rock will be tricky of course, and I assume conventional electromagnetic detection is no longer effective due to ceramic mines.

Given the accuracy of their hover capabilities, perhaps it could even have a hardened metal weight that could be dropped on the mine from 50' above and winched back up for the next target?
posted by CynicalKnight at 1:31 PM on November 20, 2012

GPR is hard to do from much height above the ground, the signal gets attenuated quickly. The systems I've seen are all vehicle or cart mounted. GPR also has a hard time with false positives. Modern mines are mostly plastic, and so tree-roots, for example, can look very similar. Dual magnetic/GPR devices, however, have promise, as do portable chemical detectors ("mechanical noses"), like IMS or portable GCMS. I think combo detectors are ultimately going to be the most successful solution.
posted by bonehead at 1:44 PM on November 20, 2012

I was expecting the simple, beautiful solution to involve the people responsible for the land mines' manufacture being marched back and forth through the minefields at gunpoint.
posted by dunkadunc at 1:49 PM on November 20, 2012

I am happy to announce my sure-fire center-pivot irrigation-based landmine detection and removal venture is now accepting capital from any and all comers.

I like that it irrigates while it clears, but is your system built to run on land that isn't perfectly flat?
posted by WhackyparseThis at 2:06 PM on November 20, 2012

GPR is hard to do from much height above the ground

Modern mines are mostly plastic, and so tree-roots, for example, can look very similar.

I think combo detectors are ultimately going to be the most successful solution.


Quadrocopter versatility to the rescue, then! Gas turbine generator powered, probably, as even lithium batteries wouldn't have the capacity for the required combination of range and lifting ability.

Mistress MineMuncher buzzes along at 10' doing a rough scan - high enough to overshoot tricky and vulnerable moving obstacles like people and animals without having to detour around them.

When a positive appears, drop and hover at 1' for a detailed scan and a good chemical sniff to confirm, then either report the co-ords to her flailing robot buddy Mr. McSlappyTank, or shoot up to 50' to drop a titanium disk on the little bugger, which she can winch back up and check for damage before continuing her survey. Every false positive, failed detonation and successful detonation can be video recorded for later evaluation by safety inspectors.
posted by CynicalKnight at 2:19 PM on November 20, 2012 [2 favorites]

Any effective system for clearing mines will generate a response by mine designers.
posted by mecran01 at 2:40 PM on November 20, 2012 [1 favorite]

If you want a picture of the future, imagine a boot stamping on a human face desert floor — forever
posted by Lanark at 4:08 PM on November 20, 2012 [1 favorite]

The quadcopters I've had experience with are all pretty small. Payloads run in the single kilo range. A GPRS unit, on the other hand, is the size of a bar fridge, or bigger, and weigh more than 100 lbs.

That's today, though. Who knows what 5 years from now will be like. I've seen systems the size of breadboxes which could be used as mag/chem detectors. IMS units have been hand-held for a decade. I would not be surprised to see small drone UAV explosive detectors in a few years.
posted by bonehead at 4:37 PM on November 20, 2012

It's too bad sound can't be used to create a big pressure wave to detonate these things. I have this science fiction story idea running through my head...
posted by BlueHorse at 5:55 PM on November 20, 2012 [1 favorite]

I'd confidently walk through any field that's been cleared after the Hero Rats have identified the mine locations.

I've been sending money to an adopted Hero Rat for a few years now. It occasionally sends me an email to let me know that it's hard at work clearing mines, and for monthly amount equivalent to a venti Frappucino, I help in my small way.
posted by jokeefe at 6:11 PM on November 20, 2012 [4 favorites]

I like that it irrigates while it clears, but is your system built to run on land that isn't perfectly flat?

It doesn't actually irrigate. Pivoting would work on non-perfectly flat land. The detection and removal bit still needs some work.

So, how many shares can I put you down for?
posted by Alvy Ampersand at 8:50 PM on November 20, 2012

In other news: Army Wants to Stop Bombs Using Halo-Style Electric Pulses
posted by homunculus at 12:01 PM on November 21, 2012

I just realized what's even worse about this design - they say it's repairable, but how on earth are you supposed to recover it after it's been blown through a minefield?

Walk over to i *BOOOOM*
posted by anthill at 10:24 AM on November 22, 2012

Another mili-irony about minefields: they don't prevent, and barely even slow down any military unit that decideds to mount an assault through that channel. Artillery, air strikes, and organic EOD people precede the tanks, then come the guys with the bayonets and machine guns. A few get blown up, but the main forces get through.

In regard to N/S Korea FEBA, the 30 or 40 thousand American troops stationed there have traditionally been known as Speed Bumps in the road of any potential North Korean assault--imagine a million screaming infantrymen running at you with bayonets. The mines will get a few. The Americans are supposed to slow the charge down until the B-52's get over the area, and turn it into a glassy field. Anyhow, that's the plan. When the dust settles, we'll focus our military-industrial patriotic fervor on avenging their brave sacrifice.

This bright idea was formulated in the early 50's, along with the IRBM theory of defending Europe: all these (intermediate-range) missiles (fired from both sides) would pass each other in the air, and land more or less in Eastern Europe, the Balkans, and so on. Naturally we assumed that Paris, Berlin, probably London, and a few dozen other large cities would be relegated into the past tense. The ICBMs were to be used in the second phase, to destroy the rest of the Soviet Union, and maybe a good part of China. They called it Mutually Assured Destruction, and thought it would be a deterrent to all-out nuclear warfare.

Parenthetical irony, non-military: You guys who are less than, say, 50 years old may wonder why your parents seem to live in a slightly different universe than you do. If so, reflect that, back in the day, this sort of foriegn policy was considered useful. So when you think about how weird minefield deployment is, consider that this notion was left over from WWI, and all these other brilliant military and political theories were built on that mentality. Your dad probably didn't notice that it was sort of twilight-zoney, and if he did, then he was (more or less) the guy watching Dr. Strangelove advise the president, and didn't have much to say about it.

We all know that we stand on the shoulders of giants. We don't usually reflect that the giants, though clever, are often nuts.
posted by mule98J at 11:27 AM on November 25, 2012

If mines are to be used, they ought to be built with a way to disarm and find them: if you transmit the right code, the explosive is disabled and the mine is ready to respond to a search device's coded ping. You would still have to be very careful about digging it up, but you'd know where it is and you'd be somewhat sure that the thing wouldn't blow you up. Combine that with overlapping systems (chemical sniffers, trained animals, etc.).

(Add a way to arm and disarm them rapidly and their owners would be able to drive over their own minefields with some confidence of not blowing up.)
posted by pracowity at 3:47 AM on December 20, 2012