Phase II Amount
$1,499,865
The Aircraft Carrier Tactical Support Center (CV-TSC) performs acoustic detection and feature extraction on passive sonobuoy sensors. Algorithms are needed to associate multiple frequency components originating from a contact, exploit signature content to improve associations across sensors and over time, and enable fusion of the characterized sonar tracks across the sonobuoy field or with radar or other non-acoustic sources. Recently, the submarine Advanced Processor Build (APB) program developed Single Sensor Association (SSA) functionality that combines all detection sources on a sensor into a single bearing-plus-frequency attribute report. Composite detections were then provided to a Multi-Sensor Association (MSA) algorithm to identify associations across sensors. SSA and MSA were independently tested and transitioned into the APB-11 baseline. Venator Solutions proposes to extend the SSA and MSA functionality to CV-TSC sonobuoys, adapting to the unique characteristics of the sonobuoy sensor. We propose extensions to the acoustic characterization to enable (a) improved SSA associations, (b) joint exploitation of acoustic association and kinematic likelihoods to improve field level fusion, and (c) evidential reasoning on acoustic and kinematic states to infer contact mode, gross behavior, and eventually contact classification. We cast the fusion process as a Probabilistic Graphical Model to motivate PGM-based evidential reasoning.
Benefit: The technology developed and demonstrated will support a path to fully automated DCL, a keystone of automated systems and reduced operator workload. The technology will be applicable to general autonomous DCL problems across all Navy surveillance, submarine, surface, and air sonar systems. The technology will also have direct application to general autonomous DCL in size, weight, and power limited UUVs, and in coastal homeland security applications. Venator will pursue military-wide use of the resulting software capability, at the direction of the Sponsor, first via the CV-TSC program, then via PEO-IWS5 AxB, PMS-485 ICP, and/or ONR PSAT programs, where the need is critical. Similar extensions should also be possible within other Services including, for example, known tracking and fusion processing for the Air Force as well as efforts known to the Sponsor. We will pursue such transitions, including joint ventures with strategic partners as appropriate for these transitions. We also expect internal research and development funding to be applied to support transitions through subcontracts or joint ventures. Commercial applications include use of intelligent reasoning and belief propagation systems, as supported by the high interest in PGM implementations. Venator will pursue applications in these areas through the appropriate agencies in a coordinated manner with the Sponsor.
Keywords: sonar automation, Acoustic association, feature aided tracking and localization, Feature association, autonomous classification, automated DCL
