N6833521C0615

Small Business Innovation Research Program (SBIR) Phase I

N211-073 - Intelligent Assistant for Anti-Submarine Warfare

Intelligent Assistant for Anti-Submarine Warfare

A typical navy sonar team faces a host of challenges in Sonar Search Planning, having to take into consideration operational profile limitations, environmental constraints and target specific considerations when deciding on optimal system search settings; settings which themselves may need to change within the course of the ASW problem as a function of variation along those three dimensions that may go unnoticed by the human operators in key parameters. These challenges are further compounded when dealing with new sonar systems, where there may not be sufficient time to build up sufficient resident knowledge on the system or leverage Navy guidance derived from lesson's learned. It is our vision to create a prototype Intelligent Assistant Tactical Decision Aid (TDA) for ASW in order to not only assist in the complex decision-making process of setting up an optimal sonar system Search Plan, but also augment the Sonar Team in continually evaluating this process. The Intelligent Assistant will utilize data from navigational and operational charts paired with environmental information from other on-board TDA's and real-time assessments in order to enable Artificial Intelligence/Machine Learning (AI/ML) algorithms to provide operators invaluable support to the ASW problem. This AI/ML algorithm is anticipated to operate in the display background as a TDA on the tactical Advance Capability Build (ACB) sonar system, agnostic of platform, and able to adapt to the system that it is operating on. By optimizing the use parameters of the systems and the recognition differential of the operators, the combined enhancements are anticipated to yield significant improvements in ASW force capabilities via seamless transition to ACB builds. While a demo for such a system can be envisioned as part of a TRL-6 PHASE II prototype, leveraging the tactical system (or surrogate) and RecFS post-processed acoustic and meta data, for PHASE I we intend to provide TRL-3 proof of concept of Intelligent Assistant capability using surrogate data and pre-existing software and/or software developed as part of the Phase I effort. For this purpose, the MAI/Clarify team propose to simulate the Sonar Search Plan decision cycle via simulated acoustic data modeled and displayed on a surrogate Active sonar display for both Pulse Active Sonar (PAS) and Continuous Active Sonar (CAS) waveforms. A range of AI/ML algorithms will then be applied to the data in order to derive potential for envisioned Intelligent Assistant to identify patterns of detection and non-detection that can both inform and adjust the Sonar Search Plan as proof of concept during the PHASE I approach. The teaming of Marine Acoustics Inc (MAI), with years of operational experience in ACB Sonar RDT&E, and Clarifai, pioneers in the development of AI/ML recognition software already in use by various DOD and Federal agencies, brings a strong marriage of capabilities well suited to this task.

The goal of phase I is to establish the technical merit, feasibility, and commercial potential of proposed R&D efforts and determine the quality of performance of the small business awardee organization.

N211-073

N211-073-1192

21.1

Ken Graf