There are many U.S. Department of Defense (DoD) sites where unexploded ordnance (UXO) cleanup is difficult because current technology is inadequate in forested areas and in challenging terrain. The man portable vector (MPV) is a handheld technology that is designed to extend the classification performance of the latest vehicle-based geophysical platforms to sites where vegetation or terrain limit access to these platforms.

The objective of this study was to demonstrate through data collection and analysis that a handheld sensor such as the MPV could be utilized for detection and classification of shallow UXO at live sites. The MPV was first validated at the Yuma Proving Ground (YPG) Standardized Test Site, Maryland and subsequently participated in the Environmental Security Technology Certification Program (ESTCP) demonstration of man-portable systems at the former Camp Beale, California which contained areas with forest and rolling terrain. The performance objectives were met by achieving high rates of detection, classification, and clutter rejection.

Technology Description

The MPV sensor head is a handheld, wide-band, time-domain, electromagnetic induction (EMI) sensor that comprises of a 50-cm diameter transmitter loop and an array of five three-dimensional receivers. This second-generation prototype was specially designed for this study to improve maneuverability and ruggedness over the initial sensor that was developed by Cold Regions Research and Engineering Laboratory (CRREL) for SERDP project MR-1443. The MPV is field programmable through a user interface that also provides immediate feedback on field data for quality control (QC) and target localization. Generally operated by a two-person team, the MPV can be operated in dynamic survey for digital geophysical mapping (DGM) and target detection, and for cued interrogation of detected anomalies.

In cued mode, high quality data are collected by taking multiple static soundings for joint interpretation through geophysical inversion. Reliable inversion requires accurate positioning. The MPV utilizes a dedicated, local, portable positioning receiver station that monitors the primary transmitter field from the MPV—like a beacon—and returns relative location estimates with cm-level accuracy out to a range of 4 m. This approach overcomes line-of-sight constraints of standard survey methods such as Global Positioning System (GPS) and roving laser rangers and, unlike a fixed template, facilitates adaptive choice of sounding locations according to target response and environmental conditions.

The MPV was deployed at two sites by Sky Research, Inc. (SKY) with technical assistance from G&G Sciences and CRREL. The data were subsequently analyzed and processed by SKY. The YPG demonstration, which occurred in October 2010, was aimed at verifying detection and classification on a blind grid after training on calibration lanes. The survey protocol involved operating in a detection search mode until a target was encountered. The detection survey was then interrupted and a cued survey, designed to collect sufficient high quality data for classification, was executed. The technology was further tested in the Desert Extreme area for its ease of use and ruggedness. The Camp Beale demonstration was conducted in June 2011 and tested classification at a live site through cued interrogation of 912 anomalies previously identified in a DGM survey using an EM-61-MKII sensor. The Camp Beale targets were split between a forested area, where vehicular access was precluded, and an open field, where performance with vehicular platforms could be compared.

Demonstration Results

High detection performance was demonstrated at the YPG site with 100% and 90% detection rates for UXO within 0.3 m and 1 m depth, respectively. Successful classification performance was also demonstrated with 95 and 100% of detected UXO (within 0.3 and 1.0 m depths respectively) correctly identified. Beacon position estimates were confirmed to be sufficiently accurate to meet classification requirements. Sensor hardware proved to be maneuverable and generally resilient. Efficient survey protocols were established and an average daily productivity rate of 100 anomaly characterizations was achieved. The system was deemed fit for live-site testing.

At the Camp Beale demonstration site, cued interrogation data were collected within the planned schedule with a daily productivity rate of 90 anomalies. To address potential issues with highly magnetic soils at the site and to mitigate anomaly location uncertainties, additional soundings were collected, thus lowering the production rate below that achieved at YPG. Data were preprocessed and made available to external data analysts for advanced processing and classification. Interpretations conducted by SKY and several other research groups resulted in the identification of 100% of the targets of interest while avoiding excavation of over 80% of the clutter. All demonstration objectives were successfully met.

Implementation Issues

The YPG and Camp Beale studies constitute the first step at establishing the performance, limitations, optimum usage, and costs of the MPV technology. The initial demonstrations were successful and showed a strong potential for detection and classification at shallow depth (up to 1 m) with a handheld technology. Of particular note is the significant reduction in the number of nonhazardous clutter items that need to be excavated: over 80% of these items could be left in the ground. Additional surveys are scheduled through 2012-2013 as part of the ESTCP Live-Site Demonstration program (ESTCP projects MR-201158 and MR-201228).

Today there is no commercially available system with similar capabilities, though there is some overlap with cart-based and man-portable prototypes that are currently being tested with the ESTCP (such as the Time Domain Electromagnetic Towed Array Detection System [TEMTADS] 2x2 and the handheld Berkeley UXO Discriminator [BUD]). The MPV sensor itself is a unique prototype. It was fabricated by G&G Sciences with U.S. government funding using “standard” components similar to those found on the Geometrics MetalMapper in addition to a number of custom features. Maintenance and integration of surveying sensors such as GPS and heading sensors currently relies on the involvement of G&G Sciences.

Technology transition efforts are ongoing. The MPV will now be field-tested without the presence of the technology developer. A wider range of surveying sensors (such as tilt meters) will be made compatible with the MPV so that legacy sensors can be replaced. All MPV data that have and will be collected in the ESTCP demonstration will continue to be made available for testing of inversion and classification algorithms and for training of analysts. Commercial partners are scheduled to participate and be trained by SKY at data collection and interpretation in upcoming field demonstrations. Demonstration reports will document survey protocols and results and will promote adoption of the MPV technology by industry.