Real-Time Data Mining and Track Fusion of National and Tactical Data

Global Maritime Operations and Naval Force Projection require an accurate and targetable depiction of the battle space.  There is a growing set of programs dedicated to the collection and collation of information, such as the Minotaur Family of Services (MFoS).  However, the proper fusion of tracks from Multi-INT has been challenging, creation of target uncertainty due to data-ringing and data-looping.  Without addressing the uncertainty issues, an authoritative target COP will not be achievable. Arete proposes to build upon their proven clustering algorithm CSSCAN, to create a software application (BACCARAT) which will provide accurately associated tracks for use within the MFoS.  In Phase I, Arete will extend the CSSCAN algorithm into a robust track fusion engine, capable of handling disparate data sets.  CSSCAN has already been shown to successfully fuse static targets in cluttered data with a high degree of accuracy on a previous SBIR.  Recently, Arete extended the approach to handle moving targets and track data.  Throughout Phase I, Arete will adapt and demonstrate the extended CSSCAN approach on available maritime data, such as AIS, and real and synthetic data as needed.  At the end of Phase I, Arete will provide a detailed report on the performance of the algorithm on the test data, as well as an initial prototype design for BACCARAT.  During the Phase I Option period, Arete will finalize plans for a surrogate system and prepare for transitioning to operational data.