Big Data Mining for Maritime Situational Awareness

Maritime situational awareness (MSA) is increasingly challenging in the ever-increasing movement of goods by shipping in global trade around the world.  Extremely dense surface contact environments are persistent in places like the Strait of Hormuz and the South China Sea including its surrounding straits where well over 1,000+ surface contacts (tracks) can be present at any one time.  Naval operations in these theaters must quickly establish and maintain the surface contact picture over long periods of time.  Current and future operating concepts (CONOPS) will employ a network centric approach that that will ingest, merge and disseminate information gathered by sensor suites to create and maintain a common operational picture.  This capability is enabled by the Minotaur Family of Services (MFoS) that associates and synchronizes the local databases across the network of users.  The primary Minotaur database is the Tactical All Source Repository (TASR) and is comprised of track objects typically consisting of multiple track attributes that can come from onboard sensors (e.g., radar, AIS, and ESM), offboard sources (e.g., sensor suites on other platforms in theater, national data, cooperative broadcast), and automatic and manual decision processes (e.g., vessel classification, operator assertions about the contacts like positive identification, etc.).  The volume of data produced by individual platforms and the network of platforms over representative mission durations could be extremely large.  As a basic building block, the proposed effort will research the application of big data mining/data analytics techniques as an aid to maritime situational awareness (MSA) at the platform/mission level for the collection of attributes that comprise track objects in TASR.