this topic may be restricted due to the technical data under US Export Control Laws.

 

OBJECTIVE: Develop advanced algorithms, machine learning, distributed computing, and/or other innovative technologies to be applied to the Search, Detect, Classify, and Localize operational phases of a torpedo mission.

 

DESCRIPTION: Undersea weapons, heavyweight and lightweight torpedoes, are launched as fire-and-forget weapon systems. A torpedo using software can be analogized to different private sector devices such as autonomous robotic vacuum cleaners; drones/robots for delivery of goods; inspection of infrastructure; and other devices, but are much more sophisticated. During the weapon pre-launch phase, the launch platform directs the device(s) to a certain target area and then the torpedo commences to search for threat targets. Analogous to this is when a robotic vacuum is directed to find dirt, avoid obstacles or other robotic vacuums, and conserve batteries. During the weapon post- launch phase, the torpedo will use various sub-phases to search, detect, track, classify, localize, target, home, and prosecute a threat target. Each of these phases can be modeled and coded as an independent Computer Software Configuration Item (CSCI). Each CSCI has its own requirements and interfaces with the other related CSCIs.

Collectively, all these CSCIs make up the torpedo operational software.

 

The torpedo functions applicable to this effort are Search, Detect, Classify, and Localize. Search defines how the torpedo processor uses data, sensors, data fusion, and statistics within areas of uncertainty to maximize the chance of finding an object while minimizing the chance an area is unsearched. Detect is the way the computer separates this data into potentially useful data versus noise. Classify is the way the computer assigns meaning to potentially useful data. Localize is the way the computer compares the potentially useful data against known values to reach a conclusion about relative positions in space.

 

The operational software application will be hosted on torpedo processor hardware that will have limited memory space and speed (timing) capability. The host torpedo hardware is expected to remain in the fleet for a minimum of 25 years and is expected to be upgradable via software updates throughout its lifecycle. Hardware processing specifications will be provided in Unclassified//For Official Use form to awardees. No Government-furnished equipment (GFE) will be required/delivered under this effort. All algorithm development should be developed on contractor-owned, general-purpose workstations using MATLAB code. Candidate algorithms should focus on detection and classification of Low-Doppler targets in countered, shallow water environments (i.e., high clutter).

 

Performance metrics for evaluation purposes will be based on the CSCI(s) affected by the proposed algorithm and will be established early during the effort. The following general criteria will apply regardless of CSCI: (1) Negative or no improvement is unacceptable; (2) An improvement of at least twenty percent (20%) in any single CSCI or ten percent (10%) in a combination of CSCIs (Search, Detect, Classify, Localize) is acceptable; (3) Government reserves the right to engage with companies that report less than acceptable performance expectations using novel approaches in the Government�s interest; and (4) Newly developed CSCIs showing favorable torpedo performance results will be utilized and provided for in-water software builds.

 

These models will be integrated by the Government with both existing and new CSCIs to evaluate weapon performance using the secure Weapon Analysis Facility (WAF) hardware-in-the-loop model and simulation environment.

 

Additional weapon capability is gained through improvement of the torpedo software through greater economy of algorithm and process efficiency. Due to commonality of software, both heavyweight and lightweight undersea weapon systems benefit from this increase; commonality will also reduce the effort associated with maintaining the software, thereby reducing total ownership cost. Also, this software can be expanded for use in Unmanned Undersea Vehicle (UUV) applications, which also represents additional capability. This effort also has the potential to increase the number of sources for torpedo software, which can reduce costs by at least 20%.

 

The Phase I effort will not require access to classified information. If need be, data of the same level of complexity as secured data will be provided to support Phase I work during the Phase I effort. The Phase II and III effort will require secure access, and NAVSEA will process the DD254 to support the contractor for personnel and facility certification for secure access.