Integration of Advanced Correlator - Navy Applications into Minotaur to Reduce Time from Detection to Classification

This sequential Phase II effort builds upon and enhances the real time, Multi INT data fusion processor called the Advanced Correlator Navy (ACOR-N), and especially some of the threat classification and identification applications developed under the original ONR Phase I and Phase II SBIR efforts. The following ACOR-N applications will be integrated into the Minotaur Family of Services (MFoS) and tested  in four tasks proposed  Task 1. Global ELINT Tracker –Enhanced (GETe): COR-N’s GET has performed exceptionally well in five Trident Warrior (TW) Experiments and in ACOR-N Harvest 2019, a coprocessing exercise with COMPACFLT in WESTPAC Areas of Interests (AoIs) for real threats. GET will be enhanced to more rapidly identify threat ELINT tracks (using Minotaur and national data sources), and GETe will be integrated into the MFoS as a real time application and tested in its ability to decrease contact detection to threat classification time. GETe will build on the first pass detection capabilities of Minotaur (eg Minotaur SLEEP mode) by leveraging Multi-Int Minotaur organic and global sensor inputs. AIS false contacts will be eliminated based on application of enhanced AREF filters. Task 2 ACOR-N ELNOT Radar Filter (AREF): AREF works with GETe like a “hand in a glove” to filter out non threat ELINT and AIS/ELINT fused threats so that threat classification, and possible identification with a national Overhead Sensor (NOS) or MQ-4C Triton IMINT. AREF has been a success in the batch processing of WESTPAC data, and will be integrated into the MFoS as a real time application, and tested with GETe in its ability to decrease contact detection to threat classification time. Task 3. ACOR-N Best Fit Algorithm (BFA): ACOR-N’s BFA has been under development since 2006, and still needs significant development and testing. BFA automatically resolvess ambiguous tracks when new sensor data is geofeasible with several existing ACOR-N tracks. This is the most difficult problem to solve for dense shipping environments. Jove Sciences Inc. (JOVE) has developed and tested 21 BFA parameters, and has recorded data on which to do the BFA evaluation. This is a critical issue to solve for the MFoS, or any data fusion processor. Jove will further develop BFA and apply national source data with its other algorithms to further resolve ambiguity. Task 4 ACOR-N’s near real time Pattern of Life Processor (PoLP): As shown in Figure 1, PoLP is a recent ACOR and development to using Machine Learning (ML) determine and database the pattern of life (PoL) of all AIS ships in the world in near real time as ACOR-N inputs tracks into the PoLP processor so these tracks can determine ”Normal Behavior” . Eventually PoLP can determine anomalous behavior when enough data has been collected. The PoLP concept will be extended to the MFoS threats of interest who generally don't broadcast AIS messages.This effort combines National and Minotaur collected data into the PoLP algorithms data base.