Central to the process of conducting mine countermeasures operations is the prosecution of underwater objects with sensors, and evaluation of the information that they provide decision-makers. This paper summarizes the methodologies, protocols, and challenges of mine countermeasures systems today, and how the same challenges affect the development of early generation unmanned underwater vehicles.

The Surf Zone Reconnaissance Project is developing sensors for small, autonomous, Underwater Bottom-crawling Vehicles. The objective is to enable small, crawling robots to autonomously detect and classify mines and obstacles on the ocean bottom in depths between 0 and 10 feet. We have identified a promising set of techniques that will exploit the electromagnetic, shape, texture, image, and vibratory- modal features of this images. During FY99 and FY00 we have worked toward refining these techniques. Signature data sets have been collected for a standard target set to facilitate the development of sensor fusion and target detection and classification algorithms. Specific behaviors, termed microbehaviors, are developed to utilize the robot's mobility to position and operate the sensors. A first generation, close-range sensor suite, composed of 5 sensors, will be completed and tested on a crawling platform in FY00, and will be further refined and demonstrated in FY01 as part of the Mine Countermeasures 6.3 core program sponsored by the Office of Naval Research.

We describe our continuing work on the development and use of collaborative virtual environments (CVEs) in support of mission rehearsal activities for the very shallow water mine countermeasures mission. These multi-participant CVE's are built from multiple data streams that include archived circulation model results, 3D model files, moving entities and observations from advanced instrumentation. A room-sized projection screen device called the CAVE^TM is used to visualize and interact with the CVE, although single walled devices and workstations may also be used. Asynchronous collaborative capability allows users at many distributed sites to partake in a many-to-many session that takes place in a common virtual world. Incorporating multiple data streams, these CVEs allow multiple users to view, navigate, and interact with data in a 3D environment, including graphical representation of bathymetry/topography, above- surface images, in-water objects (e.g. mines, bridges), and hydrographic characteristics (e.g. currents, water levels, temperature).

As opposed to deep water operations, the shallow water (littoral) areas are full of targets of all kinds, including small ships or boats of all sizes, navigational buoys and other objects with radar-reflection signatures, IR, visible- signatures, etc. Therefore, future littoral UUVs (unmanned- underwater-vehicles) should have some amphibian features and/or some similarities to UGVs (unmanned ground vehicles). In this paper, we discuss those futuristic scenarios, in the context of semi-autonomous and/or nearly-autonomous modes of operations.

An experimental dual-frequency acoustic lens sonar system, designed to detect both buried and non-buried objects is described with emphasis on the higher frequency subsystem. The lower frequency subsystem (35 - 100 kHz) forms conical beams with beam widths near 5 degree(s) using discrete transducer elements in the lens focal plane for both transmission and reception. The higher frequency (1 - 2 MHz) lens system is designed to be contained within the volume of the lower frequency subsystem to create a compact dual-frequency system. The higher frequency system consists of three 20-cm long cylindrical lenses designed to form fan-shaped beams over a 20 degree(s) field of view. The retina is positioned 34 cm from the entrance aperture. A test array containing several discrete elements with 1.0-mm pitch has been designed for initial testing. The final system will use a retina with 80 - 100 elements. The imaging system is designed to generate images with cross-range resolutions from 0.1 degree(s) to 0.25 degree(s), and is designed to be tested in both bistatic or monostatic modes. In the monostatic mode, results of spatial multiplexing of beams in the ratio of 3, 4, and 5 will be compared. The system is designed to make a thorough parametric evaluation of imaging in the 1 - 2 MHz range over a wide range of angular resolutions and to relate design parameters to operational performance for forward looking systems.

A passive sonar array referred to as `Ambient Noise Sensor' or ANS is introduced in this paper. The ANS is a simple and cost effective passive array that can provide horizontal and vertical directivity measurements and generate beamforward images with very good resolution. This array has been successfully deployed in surveillance missions to measure ambient noise due to biologics and breaking waves as well as for tracking boats, autonomous underwater vehicles (AUVs) and other sources in shallow water environment. Recent measurements for tracking AUVs that emit packets of relatively narrow band signals, however, revealed some difficulties in the ability of the array to continuously localize and track the targets. This is particularly true when other competing sources or interference are also present and their spectral content overlap with those of the targets. In this paper, a simple spectral whitening method using a large order autoregressive model is developed to smooth out the spectra of the collected signals prior to beamforming. This is then used in conjunction with the ANS correlation-based beamformer designed primarily for localizing wideband sources. The results indicate the effectiveness of the developed methods for detecting and tracking of different sources in presence of high level of interference and noise. Additionally, the ability of the proposed scheme to resolve multiple tracks associated with AUVs and boats in the beamformed images is demonstrated.

Very broadband sonar signal processing schemes have been developed in recent years for the investigation of ambient noise sources in the ocean. A new development is the application of this signal processing procedure to active sonar systems. It will be shown that good results can be obtained by using a correlation processor. This paper will describe a sonar developed for imaging bubbles below breaking waves from an AUV platform.

During the 1980's the Superconducting Gradiometer/Magnetometer Sensor was demonstrated in the Magnetic and Acoustic Detection of Mines Advanced Technology Demonstration to provide effective mine detection, localization, and classification capabilities, especially against buried mines, and to reduce significantly acoustic false alarms arising from bottom clutter. This sensor utilized Superconducting Quantum Interference Devices manufactured using the low critical temperature (low T_c) superconductor niobium and liquid helium for sensor cooling. This sensor has most recently bee integrated into the Mobile Underwater Debris Survey System and has been demonstrated successfully in a survey to locate unexploded ordnance in coastal waters.

In Phase II of a Small Business Innovation Research contract funded through the Office of Secretary of Defense, Quantum Magnetics has developed a fieldable room-temperature gradiometer (RTG). The RTG affords unprecedented dynamic range that enables detection of signals near fluxgate sensor noise (approximately 5 pT/Hz^1/2 at 1 Hz while the system is in motion in the earth's field (approximately 5 X 10⁷ pT). To achieve this sensitivity when the RTG is integrated with an autonomous underwater vehicle (AUV), the magnetic interference generated by magnetic sources onboard the AUV must be removed. Direct magnetic feedback or signal processing using both ancillary sensors and a priori information about the interfering sources can be used to recover the baseline sensitivity of the RTG.

Magnetic sensor technologies are being developed at NAVSEA, Coastal Systems Station to enhance the ability of Autonomous Underwater Vehicles (AUVs) to perform detection, localization and classification (DLC) of mines for mine countermeasures in Surf Zone/Very Shallow Water (SZ/VSW) environments. This work involves the design, development, and eventual systems integration onto AUVs of passive sensors for DLC of ferrous mines and active sensors for DLC of both non-ferrous and ferrous mines. The magnetic sensors must provide useful detection ranges and robust target localization capability for mine DLC tasks while operating aboard AUVs with challenging operational conditions of small size and power budgets, large changes in vehicle orientation in turbulent SZ/VSW areas and potentially severe electromagnetic compatibility problems. Similar technical issues must be addressed when applying magnetic sensors to solve the mission-critical problems of AUV navigation and communications in SZ/VSW environments. This paper specifically presents results from field tests of passive, DC-magnetic sensors against small ferrous targets.

Utilizing quasi-static AC magnetic fields as the channel, Magneto-Inductive Systems Limited (MISL) has developed a group of unique communication, signaling and navigation systems. We refer to this technology as magneto-inductive (MI). The physical properties of magnetic fields enable these systems to operate through any natural medium or medium boundary. Working with the U.S. Navy's Coastal Systems Station, MISL has conducted several tests and evaluations of MI system components operating in Very Shallow Water, Surf Zone and Beach Zone environments.

An SRI breadboard designed specifically for use in the underwater environment was assembled, and a second series of underwater imaging experiments were performed in the swimmer delivery vehicle test tank at Coastal Systems Station in November-December 1999. The objective of these experiments was to collect information for a critical assessment of the effect of selected system parameters on image quality, to help quantify the effective underwater performance of the SRI under controlled test conditions, and to obtain data in support of an active modeling effort. The data from these experiments are currently being analyzed. A brief description of these experiments is presented along with a discussion of selected results.

An Underwater Scannerless Range Imager illuminates a wide field-of-view with a broadbeam laser pulse and captures the entire scene with an image intensified solid state camera. By imaging the received light onto a microchannel plate (MCP) receiver whose gain is modulated, and focusing the CCD camera on the phosphor screen that fluoresces upon excitation by the MCP output pulse, a sequence of images differing in the phase of the modulation waveform can be formed, and high precision target ranges can be inferred for each pixel of the viewed scene. When the medium between transmitter and target is obscured, as by turbid water, the return signal is temporally extended so that the inferred range picks up a bias owing to backscatter. The intimate relationship between the spatial and temporal behavior of the signals (near targets produce different temporal profiles than distant ones) adds complexity that cannot be handled by point spread functions, as is common for CW illumination and range-gated systems with constant gain. The method described here breaks the propagation problem into four channels depending on whether the light is scattered by the medium on the way to or from the target (or both, or neither), and calculates arrays to represent mean pathlengths and their variances. A fairly rigorous sensor model based on the various layers in a particular implementation (photocathode, MCP, phosphor, CCD array, A/D converter) and on receiver modulation transfer characteristics completes the prescription for generating realistic synthetic USRI images in moderate turbidity.

We investigated the utility of a portable, intense source of ultraviolet light for diver use in support of Very Shallow Water operations. The working hypothesis was that the light would be of use to divers at short-to-medium ranges (up to several meters) while remaining invisible to surface observers due to the incoherent insensitivity of the human eye to ultraviolet light. The light source contained an arc discharge lamp rich in short wavelengths and was fitted with a filter that transmitted only the near ultraviolet portion of the spectrum. In-water tests were made in darkness using Navy divers both in a natural coastal environment and in a test tank. It was found that the light was of limited utility to the divers. In addition, the light was not covert because of a bluish-white glow associated with the ultraviolet beam. Subsequent measurements demonstrated that the visible glow was produced by a combination of fluorescence of dissolved organic matter in the water and Raman scatter from the water itself. The relative importance of the two factors varied with water type. These two effects that transform light from the invisible to the visible impose inherent limitations on the use of ultraviolet light for covert operations.

SEACAM is a combination of a digital camera and an underwater sonar range finder that allows the precise range to the target to be measured simultaneously with the image. Given this measured range and the known field of view of the camera, the `plate scale' of the image (i.e., the number of millimeters/pixel) can be precisely determined, allowing for accurate estimates of the target dimensions. Two prototype systems have been developed: a `functional' camera that has all the functional capability of the final camera, but which is not packaged in the final form; and an `ergonomic' prototype that represents the first attempt at the final package design, but which is non-functional.

In the shallow water military scenarios, UUVs (Unmanned Underwater Vehicles) are required to protect assets against mines, swimmers, and other underwater military objects. It would be desirable if such UUVs could autonomously see in a similar way as humans, at least, at the primary visual cortex-level. In this paper, an attempt to such a UUV system development is proposed.