A framework of real-time seabottom target detection using acoustic sensors on AUVs

The use of Autonomous Underwater Vehicles (AUV) for Mine Counter Measures (MCM) is an area of active recent research. The excellent mobility of AUV creates allows for multi-aspect sonar view of the targets for improved detection, tracking and classification. However, the capacity of underwater communication limits the amount of online data transferring. Therefore, human aided detection by images, which usually needs enormous amount data collected by high resolution sonar systems, is not feasible to online applications. A new acoustic sensing framework is proposed for detection and tracking seabottom targets based on the Track-Before-Detect (TBD) technique. In contrast to the traditional methods, the new framework uses low frequency sensors on AUVs to fully characterize the seabottom targets. There are several advantages by using this framework: 1) compared to the traditional method of using sidescan image sonars, online detection is achieved by the lower frequency sensors and thus the comparatively small amount of data makes the onboard implementation feasible. 2) Without mosaicing the high resolution images by image registry methods, the TBD overcomes the navigation uncertainty and be able to localize the targets simultaneously in the fly. 3) Without losing sufficient acoustic information of targets, the low frequency sensors can fully characterize targets in interests, especially for the buried targets, and thus, help in declaring the detection intelligently. Results of a successful application of this method in the GOATS2002 experiment are demonstrate