Many active and former military installations have ordnance ranges and training areas with adjacent waters in which unexploded ordnance (UXO) now exist due to wartime activities, dumping, and accidents. These contaminated areas include coastal waters where the adjacent lands have become further developed, and the potential public hazard from UXO encounters has continued to rise. Presently, there exists no effective capability to survey such underwater areas and map the location of buried UXO. High frequency, high resolution imaging sonars (and magnetic induction detectors) do not penetrate the sediment with sufficient signal to noise (S/N) to achieve high probability of detection and low false alarm rates for most buried UXO; and although effective against proud UXO, their coverage rates are low. Researchers have demonstrated a new class of low frequency “structural acoustic” (SA) sonars which experience little absorption in the sediment and thus can detect buried objects. In the SA domain, echoes are caused by a variety of mechanisms which generate a feature‐rich “acoustic color” spectrum that can be exploited for target classification. The objectives of this project are: (1) to demonstrate the ability of autonomous underwater vehicle (AUV)‐based SA sonars to detect and classify proud and buried UXO in a shallow water area for which the Department of Defense (DoD) holds environmental responsibility or in an area which can act as a good surrogate for such a site; and (2) to access the cost and performance of this technology.

The objective of the additional work for this project is to apply promising methods to improve the performance achieved in the blind tests. This additional work is focused on improving the classification performance beyond that achieved in the two blind tests and on adding UXO identification capabilities to the UXO versus non-UXO classification capability. This work seeks to demonstrate that advanced techniques would lead to significant improvements in the receiver operating characteristics (ROC) curves, in the ability to identify UXO type vis-a-vis classifying as a UXO, and in lowering the number of test field detections falling above the training field probability thresholds. This would lead to “dig lists’ that are more manageable in terms of the total number of detections which must be prosecuted. 

Technology Description

In SERDP Projects MR‐1513 and MR‐2103, researchers explored and successfully developed a SA‐based sonar technology for detection and classification of underwater proud and buried UXO. A principal innovation introduced in these efforts was the exploitation of acoustic color to support target classification. Further, researchers propose using a “look‐down” SA sonar implemented on a winged AUV augmented by one that is “side‐looking”. This combination provides both a wide area, high coverage rate against proud and partially buried UXO and a shorter range, high classification performance capability against buried (and proud) UXO. The look‐down configuration guarantees that deeply buried targets can be prosecuted. Further, capitalizing on the synthetic and real apertures associated with the mobile winged sonar, acoustic images can be formed with sufficient resolution to garner rough shape, sizes, and burial angles of the detected targets. The expected performance is a probability of detection/classification > 0.98 and a false alarm rate of <0.25 while operating the “look‐down” AUV sonar at a modest area coverage rate (~21m2/s or 0.5 square nautical miles/day) against buried and proud UXO and operating the “side‐ looking” AUV sonar against proud and partially buried UXO at a high coverage rate (~280m2 /s or 7 square nautical miles/day).

Skyfish AUV Sonar. Retractable sensor wings and spectrally/spatially uniform source constructed and tested near Boston Harbor.


Within the Formerly Used Defense Sites (FUDS) Program, The Army Corps of Engineers has identified over 400 underwater sites totaling at least 10 million acres potentially containing munitions, and the U.S. Navy and Marine Corps’ Munitions Response Program (MRP) has identified an additional 20 such offshore sites. Researchers estimate that more than 2 million acres (30%) involve water depths applicable to our AUV‐based technology, and a successful demonstration at an MRP or appropriate surrogate site will validate its relevance to similar sites throughout the DoD. The technology’s impact would be considerable given its unique ability to prosecute underwater buried UXO. Although the SA sonar and methodology demonstrated in this project utilize an AUV, they are also compatible with surface craft towed bodies and other platforms broadening the range of relevant DoD environments. For AUV‐based sonars, we believe that benefits to the DoD will continue to rise steadily beyond those mentioned above due to the U.S. Navy’s growing emphasis on the development and deployment of AUV system technologies. Researchers can thus capitalize on continuing advances in AUV endurance, maneuverability, and autonomy. Further, research efforts at the Naval Research Laboratory and elsewhere focused on the development of advanced onboard processing/decision making technologies would permit the development of smarter, adaptive scenarios for target prosecution based on real‐time acoustic color information and perhaps a powerful multi‐static target echo collection capability using several autonomous, coordinated and cooperating AUVs.