With permission of IMECO, International Measurement Confederation
LAMI, Ecole Polytechnique Federale de Lausanne
DeTeC (Demining Technology Center)
CH-1015 Lausanne Fax +41 21 693-5263
Invited paperThe United Nations have failed in September 1995 to ban Antipersonnel Mines. Regular armies want to keep them as a defensive weapon, and ignore the misuse guerrilla factions are doing and their dramatic effect on innocent people. Except for the metal detectors used by the demining teams, research for detecting and neutralizing mines is mostly under the control of the military and has made little progress for the last 50 years. It is time now to encourage the engineers who have developed nice control applications, personal computers and robots to be active in this field, and ask for appropriate funding. Coordination of this research must be made at the international level, and new industries must take over the manufacturing of the sensors and robots for a market of several billion dollars, if the objective is to clean the planet in 20 years. Fig 1. A sign you would not like to see in front of your door
Over the last 50 years, little progress has been made in detecting antipersonnel mines. Other presentations in this conference describe this man-made plague, which concerns 62 countries too far to really concern us. But think of the horror of living day to day in a country where at any moment you can loose a leg, or your life, or your child's life, because of these hidden, blind weapons.
About 100 millions mines have already been installed [PHR93] and are available from many manufacturers in the world [Janes95b]. For every mine painfully removed, 20 more are laid in the conflicts you see daily on CNN. Demining is slow and dangerous, with more than one accident every 5000 removed mines.
Armies use mines as a defensive weapon to protect their positions, and they keep track of their layout [Janes95a]. They develop expensive equipment for breaching through the enemy's minefields and for testing the safety of the roads at a speed greater than 20 km/h. The military are now more concerned about humanitarian demining, but they are not used to release the experience they have.
Most existing mines have been laid by guerrilla armies as an offensive weapon to spread terror, destabilize the economy, destroy food sources and create refugee flows. The low cost of mines and the difficulty to neutralize them is their key advantage. The day engineers will be able to remove a mine for a price similar to its cost (a few dollars), the strategic advantage of mines will disappear.
In the present situation, 100 million mines need to be removed for a cost of about 1000 dollars each. The annual budget for removing mines, provided by the United Nations, the governments and the NGOs (Non Government Organisations) is lower than 100 million dollars. A simple calculation shows that if no new mines are installed, 1000 years will be required to remove the existing ones. We need 10 to 50 times more money to remove them in 20 years.
The factor of 10 assumes a significant investment in research and development, while the factor of 50 is just having 50 times more deminers with their metal detectors and steel rods. With the present rate of accidents, 20,000 of them will die or be badly wounded. Reducing the mine clearing time from 1000 years to 20 years will also reduce the number of civilian casualties until the planet is cleared, from several millions down to about 100,000. We really should not accept a longer delay than these 20 years.
Year Invest- Demining R&D Indust- Cost/ Mines
ments teams ries mine removed
1996 100 M4 100 M4 0 0 1000 $ 100,000
1997 200 M$ 150 M$ 50 M$ 0 1000 $ 150,000
1998 400 M$ 300 M$ 80 M$ 20 M$ 800 $ 375,000
1999 800 M$ 600 M$ 100 M$ 100 M$ 600 $ 1 million
2000 1 B$ 800 M$ 100 M$ 100 M$ 400 $ 2 million
2001 1 B$ 900 M$ 50 M$ 50 M$ 300 $ 3 million
2002..9 idem idem idem idem decrease increase
2010 1 B$ 900 M$ 50 M$ 50 M$ 90 $ 10 million
2011..16 idem idem idem idem idem idem
Figure 2. Scenario for removing all mines before year 2020
Let us assume that for the next 3 years, the budget for demining doubles every year, and 25% of it is devoted to research, development and support to the industries providing demining equipments. Support to industries is required to help them making the initial investment, and keep the price of the technology as low as possible, assuming that even if the demining teams obtain more money, they may have difficulties to buy more expensive, new sensors and test them.
We can draw the table of figure 2, which shows that the investment in research should provide new solutions and the larger market will quickly lower the cost of demining technology. Compared to what happenend to personal computers, the figures are rather conservative; we just need to encourage small companies and not the big traditional army partners. These small companies will also know better how to adapt the technology developed for demining to other promising industrial fields, such as geological search, construction, service robots and farming.
The interesting result is that from the year 2010, the demining speed would be 100 times the present one, and the work will be finished before the year 2020. The amount of 1 billion dollars per year is not excessive for the economy of our countries, which have built many of these mines. It represents much less than 1% of the present military world budget. Distributed over the most favoured 500 million persons we are lucky to be part of, it is only 2 dollars a year. For the countries recovering their lands and being able to farm and reactivate their economy, the humanitarian help would bring a lot more than just safe walking.
Looking at the work deminers are doing, it is natural to dream about a robot carrying out this dangerous activity. Mechanical solutions (hitting the ground) are not reliable enough, with less than 95% of the mines being neutralized. In addition, these tank-looking engines are very expensive, and they destroy the water-system and the rice paddies.
As detailed by Col. Lambrecht [Lamb96], only metal detectors are used now by deminers, and the precise localization and identification of mines is made by prodding. Mines are then defused or immediately destroyed. One can consider four major steps in the demining activity:
The first step is mostly based on oral tradition (soldiers, accidents). There is the need for aerial localization of minefields by microwaves or infrared cameras, but the resuls of the expensive research carried up to now is not yet visible. Not enough research has been conducted into the detection of TNT through its influence on plants, possibly genetically manipulated. Several ideas can be worked out to make some kind of plants grow in presence of the nitrous components found in explosives. It may take one or two years before a picture could be taken from a plane, but it would be very cheap. The best case would be of course to develop an easily recognizable plant which inserts its roots inside the mine and makes it explode or allows for an easy removal.
Marking the lanes is a dangerous proce- dure that can be avoided if robots are used with a precise localization system. Some preparation of the field will also be required for robots, but the field will be larger and should be prepared in a few hours for one week of activities at least.
As for the third point, the metal detectors used nowadays detect most of the mines. All-plastic mines do exist, but there are less than 1% of these, in known places of the world. The problem with metal detectors is the amount of metallic debris in a former battlefield. Making the distinction requires new sensors, described in several papers of this and other conferences (see references). A significant software effort is required for processing the obscure and noisy data of these sensors. Sensor fusion from several different sensors will be required to increase the level of confidence. The system may require fast processors running well optimized algorithms, but we all know the progress in this field.
Concerning the last point, organizing a soccer game on the demined field is the simplest way to prove the population the field is safe. When future new sensors and demining procedures will be experimented, confidence may not exist immediately. Before the soccer game, some validation technique will have to be used. Mechanical prodding of the ground, e.g. with an arm in front of a small truck or a motorized cultivator is a solution. The next step would be to then use the same cultivator to plow and prepare the soil for the new culture. Accidents due to deep mines should not be dangerous for the farmer and easy to fix.
Researchers are used to meet in international congresses and exchange freely their experience. So do probably the military, within classified national or NATO conferences. The number of conferences on demining is increasing, with about three conferences per year (see below), plus the usual conferences on GPR, sensors, sensor fusion, or robots, which include an increasing number of papers related to mine clearance.
Research must be encouraged by all means. Most of the responsibility for this lies within the researchers themselves. Fundamental research money is available, and the application for demining may help. But one should be honest and make careful evaluations before promising revolutionary solutions. Development of new techniques will be long and painful. Their commercialization will also take a lot of effort. But it should not take more time than required; the lifes of so many civilians and deminers depend on the quality and the speed of the work.
Financial support for the development and industrialization is much more difficult to get. There is no hope for a quick return-on-investment. Applications to other fields may also take some time. The need for an efficient encouragement must be understood by the countries and the European Community. Waiting for these actions to take place, entrepreneurs have to bootstrap their activities with local financing. Many people and organization are aware of the mine problem and ready to support any serious action. An Adelaide (Australia) association decided in 1995 to fund a campaign for supporting efforts to find technical solutions to the problem of humanitarian mine clearing. Their international advertisement didn't bring any acceptable solution.
At the EPFL, the research was triggered by John Walker, an American living in Switzerland, who gave us $ 80,000 to study demining robots. Now we concen- trate on the sensors and have found an adequate support from private and public sources for building a group of 4 people, with the goal of testing a valid solution on chosen real minefields by the end of 1997 [Nic95]. We also consider important to help other researchers in this field with the information we have and to document the tests we will perform [Gros96].
Researchers should understand, and especially in this field, the importance of exchanging information and working together. Internet and the World Wide Web (addresses below) is a powerful tool for sharing all the knowledge and interests, as naive as they may be in a first step. Concerning the present major problem of data analysis and sensor fusion, files can be easily exchanged, allowing the software to be written where people are good at, and tested where sandboxes or real mine fields are available.
The different aspects of demining activities suggest many research projects. The most important are related to a safe detection and neutralization of mines [Janes95a] [Hewish95]. Potential technologies are listed in other conferences. Other projects can be very useful, and many of them have already been extensively supported by the military. Let us mention the following examples:
Robots are very attractive for many researchers. Semi-autonomous robots could save many lives, but the problem is not the same in the third world as in our countries. Police in large cities are ready to pay more than 200,000 dollars for a simple teleoperated robot just to pick dangerous objects. When deminers are paid 200 to 1000 dollars a year, and have no life insurance, the problem is different.
The low cost mine sensors we need should also be lightweight. They can be carried by a man, who has impressive motion capabilities, so many good sensors and wonderful sense of navigation. He/she is easy to program and will adapt to unexpected situations. The only advantage of robots is that they can be light enough not to trigger mines when passing over them. The robot can hence detect and mark all the mines on a field, make them blow up all together and significantly increase the productivity. The deminer may have to neutralize one mine at a time during his progression.
We should assume that the future of demining is for autonomous or semi-autonomous robots, with a strong involvement of local people. Different robot architectures should be studied and compared. Obstacle avoidance and navigation is an important problem still to be solved. Most of the solutions will be borrowed from all the work done on mobile robots. The results will have many applications for service robots, and generate a sustained market before all the mines will be removed.
Engineers and researchers have to get involved in this field, and struggle to obtain support and show their results. Mines should of course be banned, but the important duty now is to quickly clear the planet of all these mines.
[Hewish95] M.Hewish, L.Ness, "Mine-Detection Technologies", International Defense Review No10, 1995, pp 40-45
[Jane95a] "Trends in Land Mine Warfare", Jane's special report, August 1995, 161 p, cost ~$700, (info@james.com
[Jane95b] "Jane's Military Vehicles and Logistics 1995-96" 16th ed, 1995, ISBN 0 7106 1237 0, 745p, cost ~$290
[Lam96] M.Lambrechts, "AP mine clearing operations in the Third-World: experience", ISMCR'96, Brussels
[Nic95] J.D.Nicoud, "L'heritage du XXe siecle", Ingenieurs et architectes suisse, N'23, octobre 95, pp 1-8 (tires a part nicoud@epfl.ch)
[PHR93] "Landmines, A Deadly Legacy", Physician for Human Rights, New-York, 1993, 510p, ISBN 1-56432-1134
[AV/MCM95] "Autonomous Vehicles in Mine Counter-Measures Symposium", Monterey, April 1995, 690p (~$50, http://www.minwara.org/av_mcm.html)
[SPIE95] "Detection Technologies for Mines and Mine-Like Targets, SPIE Orlando Conference", April 1995, SPIE Volume 2496, 1020p (~$134 http://www.spie.org)
[SPIE96] "Detection and Remediation Technologies for Mines and Mine-Like Targets, SPIE Orlando Conference", April 1996, SPIE Volume 2765, (http://www.spie.org)
[WAPM95] "Workshop on Antipersonnel Mine Detection and Removal", Lausanne, June 1995, 74p (LAMI-EPFL, CH-1015 Lausanne, dubois@di.epfl.ch)
[FOA94] "International Workshop of Technical Experts on Ordnance Recovery and Disposal in the Framework of International Demining Operations", Stockholm, June 1994, 44p (FOA, Dept of Weapons and Protection, S-17290 Stockholm)
[CICR93] "Symposium on Anti-Personnel Mines", Montreux, April 1993, International Committee of the Red Cross, Geneva (CICR/ICRC, The Scientific Counselor, 19 Av Paix, CH-1202 Geneva)
[Sieber95] "Intnl Workshop and Study on the state of knowledge on Localisation and Identification of Anti-Personnel Mines", ISPRA, Nov. 94, Published Nov. 1995, 72p (A.Sieber, Fax +39 332 78-5469, E-mail alois.sieber@jrc.it
[Baric96] Meeting to Determine Humanitarian Mine Clearance Requirements, London, Feb 1996 (Baric Consultants, fax +44 1304 62-0082)
UXO-III|, Unexploded Ordnance Detection, Albuquerque, 27-31 mai (info:jonesmetat@aol.com)
MD'96|, Detection of abandoned landmines, Edinburgh, 7-9 octobre
(info: conference@iee.org.uk - md'96)
|Symposium on Technology and the Mine Problem - Mining, Countermine, and Humanitarian De-Mining. Monterey, Nov. 18-22, 1996
(info: http://www.minwara/org/)
Mine Facts, data base with 675 types of mines, 130 Mbytes (available also as CD-ROM), http://vricl4c.vrinet.com/infonet/minecd
United Nations: http://www.un.org/depts/landmine
Vienna Conference, September 1995: http://www.oneworld.org/landmines/
NGO Committee on Disarmament: http://www.igc.apc.org/disarm/landmine
UNICEF Anti-war agenda and statistics: http://www.unicef.org/
Stiftung Menschen gegen Minen (with many references to other WWW servers), http://www.dsk.de/mgm
MINWARA (Mine Warfare Association): http://www.minwara.org/
Minerats: http://www.fourmilab.ch/minerats/
DeTeC Demining Technology Center: http://diwww.epfl.ch/lami/detec/