Embedded environmental functional services on board autonomous underwater vehicles: Rapid Acoustic Prediction Service (RAPS)

Abstract

The use of Maritime Unmanned Systems for underwater search requires the development of software tools for embedded real-time environmental decision support to inform robotic decisions. The CMRE Multi-static Tactical Decision Aid and its web-service variant the Rapid Acoustic Prediction service have been modified to run in real time on board a dedicated embedded processor installed on CMRE's Ocean Explorer unmanned underwater vehicles. The vehicles have also been provided with a modified version of CMRE's Detection, Localization, Classification and Tracking software, the Continuous Active Sonar-Cooperative Area search INtegrated Processor, that is able to extract features that are either sensitive to, or directly measure relevant environmental, such as Direct Blast levels from bistatic sources and geo-referenced bistatic reverberation and clutter maps. MSTPA/RAPS run various hypothesis for the bottom parameters using the bathymetry, sound speed profile and sea surface roughness that are either available through service sites or can be directly measured by the support ship. The predictions for the direct blast are run over a suite of canonical bottom types and the effective bottom properties are chosen to be those which yield the best agreement between predicted and observed direct blast levels. The model can then be updated with the new bottom properties in order to create bottom scattering strength deficit maps that highlight areas of anomalous reverberation and clutter. In this paper the embedded MSTPA/RAPS and CAS-CAINPro services are used to interpret data collected during the Littoral Continuous Active Sonar 2018 sea trial conducted in November 2018 north of Elba Island.