You are here

Automated Torpedo Acoustic Classifier (ATAC)

Award Information
Agency: Department of Defense
Branch: Navy
Contract: N68335-21-C-0097
Agency Tracking Number: N191-036-0438
Amount: $1,600,000.00
Phase: Phase II
Program: SBIR
Solicitation Topic Code: N191-036
Solicitation Number: 19.1
Timeline
Solicitation Year: 2019
Award Year: 2021
Award Start Date (Proposal Award Date): 2021-01-19
Award End Date (Contract End Date): 2024-01-12
Small Business Information
11350 Random Hills Road Suite 110
Fairfax, VA 22030-1111
United States
DUNS: 784255809
HUBZone Owned: No
Woman Owned: No
Socially and Economically Disadvantaged: Yes
Principal Investigator
 Sebastian Pascarelle
 (240) 554-6323
 sam.pascarelle@indepth.com
Business Contact
 Adam Lucariello
Phone: (703) 592-0762
Email: adam.lucariello@indepth.com
Research Institution
N/A
Abstract

The Torpedo Defense Functional Segment (TDFS) is a critical component of the AN/SQQ-89A(V)15 Surface Ship Anti-Submarine Warfare Combat System responsible for protection against torpedo threats. TDFS uses hull-mounted and towed sonar sensors to detect, classify, and track threat torpedoes, provide evasive maneuver recommendations, and recommend placement of launched countermeasures. Because torpedoes are generally high-speed threats, a rapid response is required in order to execute effective evasion and countermeasure actions. In cluttered and complex underwater acoustic environments in particular, recognition of torpedo signatures is a very difficult task for any automated solution, and false alarms tend to dominate the detections. In-Depth Engineering Corporation (IEC) executed a successful Phase I SBIR effort to develop the Automated Torpedo Acoustic Classifier (ATAC) solution, an acoustic classifier for automated recognition of threat torpedo signatures to be integrated into TDFS. ATAC implements Cortical Processing – a model of sound interpretation that occurs in the primary auditory cortex of the mammalian brain – that generates a unique spectro-temporal feature set for automated acoustic signal classification using a proven machine learning solution. Using unclassified surface craft data, IEC demonstrated that ATAC can recognize a target acoustic signal (arising from “go-fast” boats) and distinguish it from other confusing noises arising from large merchant vessels, the ocean environment, and marine mammals. The proposed Phase II effort will consist of working with classified data to develop the proof-of-concept algorithms from Phase I into a full prototype system for government testing. It is expected that ATAC will be ready for Step 2 evaluation at the end of the Phase II effort.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government