ASM
Detecting Land Mines: New Technology
New ways of mapping old battlefields
The detection and accurate location of unexploded land mines, bombs or shells is a serious problem. In old war zones, mines or bombs may lie dormant for years and even decades, then suddenly go off, killing and maiming. Children, who run everywhere and touch everything, are frequent victims.
A new technique to find such unexploded ordnance (UXO) has been developed, and shows some promise. It it the result of a collaboration between researchers at the universities of New South Wales and Ohio.
The Inadequacy of Standard Approaches
The standard approach to investigating potential sites is geophysical mapping. The first phase is called 'mag and flag' in the trade. It involves screening the entire site and identifying areas for further investigation. Technicians walk along lanes, swinging a metal detector in front of them.
The most commonly used detector is a magnetometer, hence the term 'mag'. Given the nature of the exercise, a failure to detect a mine is not only messy, it is also probably the last problem that field technician will ever have. The only way to solve this problem is to make the sensor as sensitive as possible. But this means false positives, thousands of innocent points flagged as potential targets are inevitable. Another step is needed, to determine the type of each buried object before excavation begins.
The second phase normally involves an electromagnetic survey, to provide improved geophysical data over suspected areas. It is crucial that the images acquired here provide sufficient information to detect all the unexploded objects at a site. They must also be able to discriminate between hazardous and non-hazardous items, as well as identifying the type of ordnance.
About 90 per cent of the total cost of site remediation is wasted on excavating objects that pose no threat. Remediation of sites is of great concern to the US Department of Defence, particularly on its active military bases.
The Solution: Blending Technologies
The DoD's Strategic Environmental Research and Development Program is co-ordinating several efforts aimed at developing new and improved technologies to discriminate between hazardous and non-hazardous buried items.
Given the limitations of current sensor technologies, the best hope lies in detailed geographic mapping of magnetic and electromagnetic signatures. Such investigation requires geo-location technologies that function at two levels. First, anomalous signals must be coarsely located so that they can be re-acquired with an absolute accuracy of tens of centimetres. The next step is detailed mapping of signatures, which requires the measurement of the location of individual sensor readings to a relative accuracy of roughly one centimetre.
In the normal course of events, one would expect to use differential GPS. But topography or vegetation issues often mean that a site is not amenable to satellite positioning. A possible solution is to design a system based on several different technologies. This can offer a high level of absolute accuracy. It can also support reliable relative navigation accuracy in a local reference frame. The key characteristics of this approach are the level of integration, and the capability of adapting the sensor and software configuration to specific deployment scenarios.
In an initiative aimed at addressing those issues, a joint team, from the Ohio State University's Satellite Positioning and Inertial Navigation Laboratory and the University of New South Wales' School of Surveying and Spatial Information Systems, is working to develop a geo-location system. It is an integration of four technologies: GPS, INS (Inertial Navigation System), a terrestrial RF (Radio Frequency) system, often called a pseudolite, and terrestrial laser scanning.
The team includes Professor Dorota Grejner-Brzezinska, Dr Charles Toth, Dr Hongxing Sun and Xiankun Wang, all from Ohio State University, and Professor Chris Rizos from UNSW.
The collaboration is now 14 years old. Professor Grejner-Brzezinska said the UNSW group has a very strong background in signal processing and hardware development. The laboratory at Ohio State is strong on algorithm design for integrated systems, she noted, having worked on many software implementation projects for various government projects that were focused on precision navigation and sensor integration. She said that in that sense, the collaboration was an ideal 'marriage'.
The Mechanics of Integration
The objective is to develop a high accuracy and high reliability system based on multi-sensor integration. It will significantly improve the state of the art in sensor georegistration.
Tight integration of GPS and INS has been readily available for many years. However, the integration of pseudolites and laser scanning is novel. The key aspect of the Ohio State component is terrestrial laser scanning, while UNSW has contributed its pseudolite expertise.
The pseudolite technology uses the Locata system. Pseudolites were originally conceived as groundbased substitutes for GPS Navstar satellites, which could be designed to mimic the signals from Navstars in situations where the satellite signals were inadequate for one reason or another. This would make it possible to use standard GPS receivers with signals generated from pseudolites, mixing them with satellite-based signals. But it has proved extremely difficult to design a system that can do this in practice.
So Locata has developed a dedicated system. It is based on a network of dual frequency ground-based transmitters that cover a survey area with strong signals of continuous coverage. It has been under development in Canberra for several years. It involves a network of reference stations called LocataLites, which create a LocataNet. These signals transmit in the 2.4 GHz industrial, scientific and medical band. The receiver uses four or more ranging signals to compute a high-accuracy position completely independent of GPS. Locata allows complete control over the ground-based constellation, leading to optimal positioning geometry and consistent centimetre-level accuracy.
The laser scanner is designed to produce automated high-speed data capture of complex surfaces, often in inaccessible environments. Its accuracy can range from sub-millimetre on small object scans, to 25 mm on objects at distances up to 250 metres. Absolute positioning can be obtained by using targets attached to fixed reference locations.
A Hierarchical Approach
The team adopted a hierarchical approach in the problem. At the top is an absolute or global solution. It is available in open areas, and is achieved with GPS. GPS is the source of the Earth Centred, Earth Fixed co-ordinates. Then INS, calibrated by GPS, provides the attitude angles of the geophysical sensors.
The second level provides a relative medium range solution. This is suitable where GPS signals cannot be easily acquired. Locata substitutes for the GPS, so Locata/INS will be used to make the connection between areas with good GPS reception and those with limited GPS. Locata will substitute for GPS signals for medium transmitter/ receiver separation.
At the third level, the relative shortterm solution, very high relative accuracy is required. This will be achieved using lasers in a local reference frame. The laser scanner can be connected to the GPS/INS/Locata system, thus maintaining absolute positioning. The relative positioning accuracy in the local frame will be at centimetre level.
Grejner-Brzezinska said: 'Even though each sensor can work separately and provide some components or a full navigation solution, ideally they should work in synchronisation. Assuming that they are available, GPS and Locata must calibrate the INS, which takes over the navigation process when the RF systems are blocked or jammed. TLS can work entirely independently, but it is only capable of providing the change in position. So to navigate in a global frame rather than in a local system, it must have a starting position derived from GPS.
'Ideally, all sensors are integrated in a so-called tight integration mode, where all of them work in synch. They deliver streams of measurements to a single extended Kalman filter, which then produces a navigation solution. 'It requires that all sensors be placed on the same platform. That solution has presently been implemented in our prototype.'
Sufficient research has already been done to know that the tight integration of these four diffeent technolgies can be achieved. The real issue still to be determined is whether or not the accurate identification of electromagnetic signatures leads to reliable detection. The most important question, then, will be whether this leads to a method that is cheap enough to make a difference.
The Ottawa Convention
The best solution to the problem of unexploded ordnance would be to stop placing it in the ground at all. A movement to ban its use began in the late 1980s and culminated in the Ottawa Convention of 1997, the Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti- Personnel Mines and on Their Destruction. The treaty is the most comprehensive international instrument for ridding the world of the scourge of mines and deals with everything from mine use, production and trade, to victim assistance, mine clearance and stockpile destruction.
In December 1997 a total of 122 governments signed the treaty in Ottawa, Canada. In September the following year, Burkina Faso was the 40th country to ratify, triggering entry into force six months later. Consequently, in March 1999, the treaty became binding under international law, and did so more quickly than any treaty of its kind in history. Today, the treaty is still open for ratification by signatories and for accession by those that did not sign before March 1999.
As of March 2008, there are 156 member states and 39 states that remain outside the treaty, including two signatories that have not yet ratified.
Mines are still produced in 13 countries. Nine of the 13 mine producers are in Asia (Burma, China, India, Nepal, North Korea, South Korea, Pakistan, Singapore, and Vietnam), one in the Middle East (Iran), two in the Americas (Cuba and the United States), and one in Europe (Russia).
