Intelligent Machines

New Hope in the Minefields

Emerging technologies can speed the removal of the millions of burried landmines that continue to kill and maim civilians in more than 60 countries.

Along a path at the edge of a weed-ridden farm in Cambodia, a man listens carefully as he sweeps his metal detector over the ground. When the detector’s squealing tone signals the presence of buried metal, the man stops, repeats the sweep, and carefully marks the spot. Soon a second worker follows and lies on the ground, his head an arm’s length from the marked spot, gently probing the ground with a stick. Both men are experienced deminers, one a retired British Army veteran far from home, the other a local resident trained to find mines. Both know well the cost of error: sudden serious injury or death.

After probing the hard dirt with concentrated care for about 20 minutes, the prone worker judges by sight and feel whether he has hit the rounded metallic body of a buried mine or merely the random detritus of an old battlefield: a bullet, a piece of shrapnel, a length of wire, an empty tin can. He knows that in some fields the odds are as low as one in a few hundred that the detected metal is actually a mine, but his partner’s metal detector cannot distinguish an explosive device from any other object that holds a fraction of an ounce of metal.

Whatever it is, the metallic object must be carefully exposed to reveal its form and color. If it is a mine the workers will place a modest explosive charge beside it, unspool long wires, and retreat 100 yards to blow it up. Then the task will repeat itself: the operator of the metal detector will resume his patient sweeping, listening for signs of the next buried metallic object in his path while his partner waits for his next tense trial in the dirt.

This story is part of our October 1997 Issue
See the rest of the issue

According to United Nations estimates, more than 100 million mines lie buried around the world, outlasting their wars, abandoned long ago yet awaiting their unintended victims for as long as decades. An anti-personnel mine costs only a few dollars to produce, but it now costs a hundred times that sum to remove it. In Cambodia alone, where some of the world’s densest minefields lay, roughly 10 million mines lurk within an area the size of Missouri. Last year three thousand workers cleared landmines from 12 square kilometers of Cambodian land at a cost of $8 million. They were not overpaid. But at that rate, even if someone were willing to foot the bill, demining Cambodia would take some 10,000 years. To make matters worse, participants in today’s conflicts are emplacing new mines at a rate 10 or more times the current yield of the deminers, who now clear perhaps 100,000 mines per year worldwide. A chronic and growing crisis is at hand.

Most poignant is the human toll that the residual landmines claim: some 10,000 deaths annually and at least twice that many serious injuries, with victims including many small children and elderly villagers in poor nations. In Cambodia, landmine accidents have resulted in one amputee per 250 people. Yet clearing away lingering landmines is not needed just to protect human life and limb. Over the long term, landmines disrupt normal economic activities such as travel and transport, and deny vital cropland to farmers, often causing hunger and forcing sizable agrarian populations to migrate to urban centers and refugee camps.

Today, the use of a metal detector, hand-held probe, and explosive charge is generally accepted as the most reliable demining method despite its laborious and perilous nature. The detection method works because most mines have metallic casings or at least contain a few grams of metal, usually a firing pin and its associated spring, setting off a signal in the detector even when a mine is buried or hidden beneath overgrown vegetation.

The bottleneck occurs, however, in discriminating between the few real mines and the many false alarms. Given the wide array of metal objects that can reside in the soil of former battlefields, the false-alarm rate can run as high as 1,000 false positives to one real mine. The result is that the bulk of the searcher’s time is spent on the painstaking exposure of harmless metal scraps. And after hundreds of false alarms, the job becomes even more perilous: one surprise mine can maim or kill deminers whose patience has flagged just once, causing them to misjudge the form they uncover. What’s more, growing use of plastic-encased mines poses the ominous threat of false negatives: that real mines will remain silent-and deadly-even when swept by a metal detector.

Despite the admittedly grim situation, though, we find some cause for optimism since reviewing the global landmine problem at a week-long meeting last summer. Conducted by the MIT Program in Science and Technology for International Security at the American Academy of Arts and Sciences in Cambridge, Mass., the meeting drew together a disparate group of participants, including a field worker from Laos with many years of demining experience; researchers with expertise in physics, chemistry, electrical engineering, material science, and anthropology; several people working on high-tech mine-detection schemes; and three experts on demining from the military who brought the group a sobering collection of anti-personnel mines (without explosives). Our unexpectedly hopeful view, bolstered by subsequent study, is that while no silver bullet appears to be on the near horizon to solve the demining problem, promising technologies at hand can offer significant help. A number of developing techniques, for instance, detect landmines by sensing physical and chemical properties other than metal content, thereby significantly aiding in the task of reliably discriminating mines from metal scrap. Our analysis indicates that if nations lend enough support, affordable technologies could be available in the field within five years to undertake a humanitarian demining effort on an unprecedented global scale.

A Primer on Landmines and Their Removal

Some 700 different models of mines can be found worldwide. Designs differ widely, especially among those mines developed over the past 20 years. The most common landmines are the millions made for use by the militaries of such big powers as the former Soviet Union, China, and the United States and sold around the world. More than a dozen industrialized countries, including Czechoslovakia, France, Italy, and Yugoslavia, have also produced and sold or given away significant numbers of mines.

The major practical distinction among different types of landmines is their intended target. Mines big enough to destroy vehicles are known as anti-tank mines. These mines, roughly the size of large stove-top pots and pans, contain 10 pounds or more of high explosive. Considerably more prevalent, anti-personnel mines are roughly the size of cans of tuna. Containing anywhere from less than an ounce to a half-pound or more of high explosive, they are designed to maim or kill individuals or small groups on foot.

Mines also differ in the cruel cunning of their designs. Sophisticated mines of all sizes may, for instance, incorporate countermeasures against demining. Some, employing an accordion-like trigger design, can withstand the sudden shock of a nearby explosion, detonating only when more slowly depressed, as by the pressure of a foot; others employ anti-disturbance devices that detonate the mine whenever it is handled, injuring or killing would-be deminers. Bounding mines spring up three feet above the ground to shatter into fragments with a lethal radius of 90 feet. And some larger mines may even emit directed fragments: the large U.S. Claymore mine used in Vietnam, for instance, has a 150-foot lethal range for persons walking into its line of fire.

Because the larger-size anti-tank mines cost more to produce and lay, they are much less numerous, increasingly more sophisticated, and generally found on roads or around military installations and other centers of travel and communication. By contrast, anti-personnel mines are cheap, numerous, and prevalent in many diverse locales. The damage anti-personnel mines inflict-disabling victims for months or for life-is economically worth orders of magnitude more than their cost of a few dollars apiece. By that cruel calculus, they are cost-effective even against irregular infantry or the poorest of unarmed villagers. Because of their prevalence and availability, because they tend to be placed more randomly, and because they make up the bulk of the lingering scourge, these anti-personnel mines are our quarry, the particular focus of humanitarian demining efforts.

To be sure, mines are not new weapons and armies have long developed methods and organizations for demining. But the fact is, today’s task of large-scale humanitarian demining is new, and not really open to swift solution by deploying military-trained combat engineers. Humanitarian demining entails peacetime detection and deactivation, over an indefinite period of time, of virtually every mine emplaced in a wide area-a place of home and work to many people whose resources are often scarce and life arduous. Humanitarian demining demands nearly 100 percent detection. The search can be very slow, large numbers of false alarms are acceptable even though costly, and all operations can be confined to good weather and daytime conditions. With these dramatically differing requirements, it is not surprising that demining methods and equipment vary widely.

By contrast, most military demining efforts have favored a capital-costly “brute force” approach that uses motorized vehicles equipped with steel rollers or treads able to detonate anti-personnel mines by riding over them, with damage to the vehicles minimized through clever design and heavy shielding. Some are heavily armored trucks that ride roughshod over mines withstanding most of the anti-personnel detonations with only minor and largely reparable damage. Others, like big bulldozers, attempt to pick up and remove mines, clearing a path as they go.

Such vehicles are particularly suited to so-called military tactical demining which aims to “breach” minefields, rapidly clearing corridors, paths, and roads for combat use even during battle, often within hours. But the brute force approach is largely inappropriate for the highly exacting task of humanitarian demining: when it is applied to uneven ground, it may not detonate every explosive device. Yet such an assurance is precisely what local inhabitants need. The customary test of demining success is direct and public: as the neighborhood watches, the deminers join hands to form a line and walk across the entire plot. Would you yourself settle for less?

Unfortunately, the great variety of fusing mechanisms, of emplacement methods, and of terrain makes the thorough neutralization of anti-personnel mines decidedly difficult. While unquestionably heroic and well suited to the world of low technology, the present creep-and-probe method of humanitarian demining is plainly unaffordably slow, expensive, and dangerous. Because of these drawbacks, creep and probe demining as it is currently practiced can have only marginal impact on the global landmine problem. A true solution mandates developing and quickly deploying new methods and equipment that can speed up humanitarian demining by up to a hundred-fold at affordable cost.

Improving Creep and Probe

We believe three currently available technologies, when used together, could offer a 10- to 20-fold improvement over today’s demining rate within the next two years. These technologies include detection by a variant of the electronic metal detector (called the meandering winding magnetometer); safe and swift excavation by a device called an air knife; and detonation by a cheap and easily deployed foam-like explosive. All three of these improved demining technologies still require field testing and refinement, but the development tasks look modest.

The basic operating principle of the new meandering winding magnetometer (MWM) detector is the same as that of conventional metal detectors that use a pulsed-electromagnetic induction sensor. But whereas conventional detectors generate an electromagnetic field and sense if it is disturbed by conducting material in their path, MWM detectors generate a varying magnetic field that excites currents in metallic objects that align primarily in one direction and can be read by the detector. An MWM detector slightly larger than a conventional metal detector can thus obtain a crude hint of the size and shape of a buried metallic object by combining readouts of these so-called eddy currents. The MWM detector now being developed by Jentek Sensors Inc., of Brighton, Mass., can reportedly determine the rough size, shape, depth of burial, and type of material of the outer shell of a buried metal object. Laboratory evidence indicates that the device can provide enough information for an experienced operator to discern whether a buried object is mere clutter, a mine, or a larger piece of unexploded ordnance.

Field tests of a first-generation MWM prototype indicate that it can lower the false-alarm rate by a factor of 5 to 10 below that of a conventional metal detector. Given such discriminating power, a refined version of such an MWM device could reduce the time spent examining a square meter of scrap-rich ground from 10 to 20 minutes to a fraction of a minute.

Once a mine is detected, the air knife, now commercially available although not in field-ready form, offers a significant improvement in efficiency and safety over the stick commonly used in today’s demining efforts. The air knife blows high-pressure air through a small hand-held probe and can blow away most dirt to expose mines without disturbing them enough to detonate them. Existing air knives are powered by a 3-horsepower gasoline engine, like those that run power lawnmowers, and cost a few thousand dollars. A version adapted for demining could replace the simplest manual probes, greatly speeding up a searcher’s ability to expose a mine while improving safety at the same time by obviating the need to dig in the ground with a stick.

The use of the product Lexfoam will also aid demining efforts. The product, much like shaving cream in appearance, is a dilute dispersion of an explosive contained within a foaming plastic substance. Lexfoam is safe and simple to apply and can be set off by an ordinary detonation cap, removing the delicate and hazardous task of wiring a charge onto an unearthed mine. We estimate the use of such a product to blow up the exposed mine would considerably speed up the overall demining process, perhaps by as much as a factor of 2 to 5.

The air knife would require an air compressor (or compressed air supply) carried on a hand-drawn wheeled cart, packaged into a backpack-like portable unit, or built into a small motorized vehicle that carries the MWM metal detector, air knife, and Lexfoam dispenser. In a small, relatively new humanitarian demining unit at Fort Belvoir in Virginia, the U.S. Army is now assembling such a vehicle that combines an MWM detector, compressor, air knife, an operator’s plastic shield to protect against explosion, an air-operated weed-cutter, and a Lexfoam dispenser. Col. Harry (Hap) Hambric, who directs development and testing in this unit, estimates that the combined use of these relatively simple technologies where terrain is suitable could speed up demining by a factor of 10 within a year or two, and another factor of 10 with refinements to come.

Just as creep-and-probe methods can find quick technological improvements, though, the brute force demining of open spaces, like field and paddy, can also profit almost at once by the adoption of simple technical improvements. One promising approach proposes to use a small-sized tined roller with hinged spring-loaded prods that can set off anti-personnel mines as it passes over them. The rope-towed (or winched) roller is simple, inexpensive, and easy to repair. It contains hundreds of closely spaced, stiff, spring-mounted fingers able to penetrate up to 25 centimeters into the ground; the roller is towed back and forth across the target area using power supplied by animals or motor vehicles kept at a safe distance.

Tests under controlled conditions performed by the U.S. Army at Fort Belvoir in 1995 proved that the roller was capable of exploding or otherwise destroying small anti-personnel mines even in the mud bottom of rice paddies and other soft floored terrain. A footpath-sized version of the roller also proved to be easily repaired using simple hand tools and hardware. The roller was effective against mines in soft ground and mud. With some design modifications it could be configured to operate on harder surfaces, including areas bearing light foliage.

Thus in certain terrain this technology will allow the welcome option of clearing anti-personnel mines without detecting them first. The Fort Belvoir experts estimate the cost of this multi-pronged roller to be under $20,000, adding that it could drop to as low as $5,000 if the device were mass-produced. The group hopes to field test the system shortly. Taking tools of this sort into the field-even these initial aids imply further improvements-will make a large difference at whatever scale they are put to work. The whole job cannot be finished soon; indeed, a long-lasting culture of understanding and vigilance in the whole countryside, and a reliable source of technical aid from beyond the village-including personnel, equipment, and training-will have to be established in the most affected countries. Determination to keep up and extend the good work will thrive if visible progress comes soon in one or two places.

High-Tech Detection

While near-term technological improvements offer hope for better demining efficiency, technologies undergoing vigorous research and development for use against airline terrorism offer even more promise for the future. Portable, rugged versions of these technologies, which detect small amounts of explosives, would be required for use in demining, but the task is certainly not beyond the capabilities of high-tech firms in the United States and elsewhere.

These technologies could take advantage of the fact that landmines use characteristic materials in well-defined shapes and sizes, giving them mechanical, acoustic, electromagnetic, and nuclear absorption and reflection properties potentially detectable from a modest distance. All mines contain high explosives, substances otherwise rare in the soil, and are thus open to many means of detection based on their chemical composition.

Such chemical sensing is perhaps the most advanced of these avenues. Since all mines contain 10 grams or more of explosives, one way to avoid the time-consuming step of discriminating mines from false alarms, and to detect plastic as well as metallic mines, is to devise detectors sensitive to the presence of explosives, either in their condensed or vapor phase. We know that mines carry traces of their explosives because dogs trained to scent high explosives can detect buried mines under field conditions in a short time, with a 95 percent success rate and a false alarm rate of around two to one. Unfortunately, though, dogs tire easily and are expensive to train and keep. Arrays of sensors, each with some specificity to a particular molecule or compound, are quite commonly used in the food and perfume industries to identify products’ constituent compounds. The U.S. Defense Advanced Research Projects Agency is actively pursuing an array of such sensors intended for explosives detection at airports that could well be adaptable for humanitarian demining.

One detector already in trial use at airports pulls an air sample through to a collector that transfers any minute traces of explosives to a separation device. There, an instrument called a high-speed gas chromatograph separates explosives from one another and from non-explosive compounds by the length of time it takes each compound to go through the instrument. Each compound yields a reliable and characteristic signature. Noting both this signature time and its amplitude, the detector can determine the type of explosive and the level of its concentration in the air sample. The manufacturer, Thermetics Detection based in Woburn, Mass., claims that its system can detect the presence of 10 to 20 picograms of TNT-a grain twice the size of a speck of dust-with a thousand times the sensitivity of a dog. The system is capable of detecting picogram levels of explosives in less than a minute, and has worked well in the presence of potentially interfering compounds in the air or the soil.

Company representatives believe that a single portable, battery-powered detector could detect mines with greater than 90 percent accuracy while scanning ten square yards per minute. What is not known yet is to what degree high-explosive vapor and particles deposited by past weapons firing in the areas where mines are buried might generate an unmanageably high level of background noise. Detailed field measurements at the sites of past combat, as well as of background levels in battle-free and mine-free areas, must be conducted before the practicality of this potential mine detector can be fully determined.

At least two other technologies could potentially be used to detect mines by sensing their main charges. The first is based on nuclear quadrupole resonance (NQR), an electrostatic relative of magnetic resonance imaging now familiar in the medical world. NQR is an effect displayed by atomic nuclei that are not spherically symmetrical but slightly squashed or elongated at the poles. Nitrogen atoms, a near-universal primary ingredient of high explosives, possess just such nuclear asymmetry. Depending on what kind of crystalline structure the nitrogen nuclei find themselves in, their non-sphericity produces a unique set of very narrowly spaced energy levels that is characteristic of the crystalline solid itself. An explosive compound can therefore be identified by the subtle resonance of its constituent nitrogen atoms.

NQR detectors have already been tested in airports, where they have managed, within six seconds, to detect the military explosive RDX in quantities comparable to those in a mine. Tests at the Naval Research Laboratory based in Alexandria, Va., have shown that NQR detectors, unaffected by soil contaminants like metals and magnets, can reliably discern explosives from other nitrogenous materials in the soil such as fertilizer or living organisms. A field NQR detector would operate much like a hand-held metal detector but would require a backpack to accommodate its larger battery. NQR commercial developer Quantum Magnetics of San Diego estimates that a prototype mine detector based on NQR could be developed within two years at a cost of about $1 million. The price of such detectors, once produced in quantities of several thousand, they believe, would probably be under $10,000 each-some two to three times more than the cost of high-quality metal detectors. With an adequate level of development funding, it is quite possible that NQR could become an effective tool for discriminating mines from metal clutter within 3 to 5 years, reducing the false alarm rate to negligible levels.

The technology does pose some difficulties at present, however. The dominant obstacle is the efficient detection of TNT, the explosive ingredient of 80 percent of landmines. TNT has an intrinsically weak NQR signal, requiring a longer integration time in the detector. An NQR mine detector that had to linger over each spot on the ground for minutes at a time would clearly be too slow, although it could still presumably prove useful in distinguishing mines from scrap metal.

A second way to detect plastic mines by their explosive content is to “illuminate” the ground with a beam of low-energy x-rays. Because of the difference in their average atomic numbers, soil will absorb low-energy x-rays impinging upon it, while the lighter mine will “backscatter” a large fraction of the incoming radiation. When imaged, the mine thus appears as a luminous spot on a dark background of soil. Experiments conducted as early as 1975 by the U.S. Army Mobility Equipment Research and Development Center showed that, while awkward, clumsy, and dangerous at the time, the method does in fact work, unambiguously detecting small (six centimeters in diameter) plastic mines buried under two centimeters of soil.

Although x-ray backscattering detectors perform well in detecting explosives and other materials with low atomic numbers at airports and customs stations, they have shortcomings for detecting plastic mines: they cannot reliably discriminate explosives from other materials with similar atomic numbers (such as roots and water), they detect only shallow-buried mines, and they require an intense source of ionizing radiation that could cause health hazards to the operator. A hand-held detector may therefore not prove practical, but x-ray backscattering detectors might eventually be used on remotely controlled demining vehicles to detect plastic mines in conjunction with a metal detector such as the meandering winding magnetometer.

Launching a Global Demining Campaign

To make humanitarian demining feasible on a large scale, the demining rate must greatly increase to reduce overall costs and justify expenditures for more sophisticated equipment. This will require a gradual shift from a labor-intensive low-tech approach to the intermediate stage of introducing power tools and discriminating detectors. The final stage will require the development of autonomous, mechanized demining systems to incorporate some of the more sophisticated detection technologies we have described. Such progress will require a coherent, sustained, and adequately supported R&D effort in the range of tens of millions of dollars annually over several years.

Unfortunately, frustration with the marginal results of even the most heroic demining efforts so far has led to a tired indifference among the public and decision makers alike. This frustration has, in turn, led to the loss of opportunities for new solutions. The constellation of humanitarian relief organizations that have patiently shouldered most of the demining efforts, including the Red Cross, CARE, and the United Nations, to name just a few, have had little contact with the scientific and technical communities in academia and in high-tech industries that could boost demining efficiency. For their part, the scientific and technical communities in the developed world have largely ignored the problem. As an example of this lack of technological partnership, a one-million-pound reward offered several years ago by the British Government for an acceptable plan to demine the remote and difficult terrain of the Falkland Islands has gone unclaimed.

The latest Insider Conversation is live! Listen to the story behind the story.

Subscribe today
Already a Premium subscriber? Log in.

Uh oh–you've read all of your free articles for this month.

Insider Premium
$179.95/yr US PRICE

More from Intelligent Machines

Artificial intelligence and robots are transforming how we work and live.

Want more award-winning journalism? Subscribe and become an Insider.
  • Insider Premium {! insider.prices.premium !}*

    {! insider.display.menuOptionsLabel !}

    Our award winning magazine, unlimited access to our story archive, special discounts to MIT Technology Review Events, and exclusive content.

    See details+

    What's Included

    Bimonthly home delivery and unlimited 24/7 access to MIT Technology Review’s website.

    The Download. Our daily newsletter of what's important in technology and innovation.

    Access to the Magazine archive. Over 24,000 articles going back to 1899 at your fingertips.

    Special Discounts to select partner offerings

    Discount to MIT Technology Review events

    Ad-free web experience

    First Look. Exclusive early access to stories.

    Insider Conversations. Listen in as our editors talk to innovators from around the world.

  • Insider Plus {! !}* Best Value

    {! insider.display.menuOptionsLabel !}

    Everything included in Insider Basic, plus ad-free web experience, select discounts to partner offerings and MIT Technology Review events

    See details+

    What's Included

    Bimonthly home delivery and unlimited 24/7 access to MIT Technology Review’s website.

    The Download. Our daily newsletter of what's important in technology and innovation.

    Access to the Magazine archive. Over 24,000 articles going back to 1899 at your fingertips.

    Special Discounts to select partner offerings

    Discount to MIT Technology Review events

    Ad-free web experience

  • Insider Basic {! insider.prices.basic !}*

    {! insider.display.menuOptionsLabel !}

    Six issues of our award winning magazine and daily delivery of The Download, our newsletter of what’s important in technology and innovation.

    See details+

    What's Included

    Bimonthly home delivery and unlimited 24/7 access to MIT Technology Review’s website.

    The Download. Our daily newsletter of what's important in technology and innovation.

You've read all of your free articles this month. This is your last free article this month. You've read of free articles this month. or  for unlimited online access.