N113-176 TITLE: Multi-Target High Probability of Kill Weapons Engagement

TECHNOLOGY AREAS: Information Systems, Sensors, Weapons

ACQUISITION PROGRAM: Undersea Defensive Warfare Systems Program Office (PMS 415). ACAT III

RESTRICTION ON PERFORMANCE BY FOREIGN CITIZENS (i.e., those holding non-U.S. Passports): This topic is "ITAR Restricted". The information and materials provided pursuant to or resulting from this topic are restricted under the International Traffic in Arms Regulations (ITAR), 22 CFR Parts 120 - 130, which control the export of defense-related material and services, including the export of sensitive technical data. Foreign Citizens may perform work under an award resulting from this topic only if they hold the "Permanent Resident Card", or are designated as "Protected Individuals" as defined by 8 U.S.C. 1324b(a)(3). If a proposal for this topic contains participation by a foreign citizen who is not in one of the above two categories, the proposal will be rejected.

OBJECTIVE: The objective of this SBIR is to optimize fire control through innovative research and development in machine cognitive decision theory to develop a fire control decision engine that addresses the complexities associated with the simultaneous engagement of multiple concurrent hostile torpedoes while addressing the uncertainty dimensions and associated constraints.

DESCRIPTION: The Torpedo Warning System is a man-in-the-loop system that couples active and passive sonar components with a fire control decision engine to engage incoming torpedoes with CATs. The man-in-the-loop role is to apply situation awareness using a clear and simple information display to validate automated torpedo alerts and to make decisions concerning launch of CATs and ship�s evasive maneuvers. The actual fire control guidance to optimize CAT effectiveness is automated. Current program-of-record fire control solutions are built upon an explicit enumeration of inputs and behaviors where system designers attempt to anticipate all possible behaviors of the system. This solution provides a base capability that is repeatable and auditable, but not robust in the entire solution space.

Recent academic developments in the area of adaptive machine learning have not been applied in this arena. This SBIR seeks research only in the application of Adaptive Learning techniques to the TWS multi-target problem. Machine learning systems adaptively improve with exposure to the problem space. Evolutionary algorithms, genetic programs, classical neural networks, spiking nets, and learning classifier systems seem suitable to address this problem. This topic does not seek development of all the technologies mentioned above but does seek the application of one or more of these implicit techniques to the Torpedo Warning System (TWS) problem that is measurably superior to the program-of-record approach. Small businesses will utilize modeled or simulated data based upon publicly available information to develop the Adaptive Learning approach through phase II. Given that learning systems provide limited auditability, the proposed solution must prove to be deterministic in the sense that, once deployed, the behavior in a given set of circumstances must always be the same (repeatable).

PHASE I: Develop criteria concepts to discriminate amongst modern machine learning approaches with applicability to Torpedo Warning System (TWS). Provide recommended approach/design for prototype system with Phase II program plan.

PHASE II: Develop prototype machine learning system based upon results of Phase I, using simulated data. Develop Metrics and assess relative performance of learning system against explicit enumerated system.

PHASE III: Provide development of a scalable system with interfaces to Torpedo Warning System (TWS) and implement the recommended system developed under Phase II. Evaluate and demonstrate the system�s ability to augment the Torpedo Warning System (TWS).

PRIVATE SECTOR COMMERCIAL POTENTIAL/DUAL-USE APPLICATIONS: Advances in machine cognitive decision theory are applicable to automation efforts going on in commercial rail industry, automobile automation programs, robotics industry, as well as the commercial power industry.

REFERENCES:
1. Marsland, Stephen (2009), Machine Learning: An Algorithmic Perspective. Chapman & Hall/Crc Machine Learning & Pattern Recognition

2. Bishop, Christopher (2007), Pattern Recognition and Machine Learning. Springer, Corr. 2nd printing edition

3. Winkler, Joab; Lawrence, Neil; Niranjan, Mahesan (Eds.) (2004), Deterministic and Statistical Methods in Machine Learning. Springer Lecture Notes in Artificial Intelligence

KEYWORDS: Machine Learning; Cognitive Decision Making; Human-Machine-Interface; Defensive Warfare Systems; Visualization; Rapid Response

** TOPIC AUTHOR (TPOC) **

DoD Notice:   Between July 28 and August 28, 2011, you may talk directly with the Topic Authors to ask technical questions about the topics. Their contact information is listed above. For reasons of competitive fairness, direct communication between proposers and topic authors is not allowed starting August 29, 2011, when DoD begins accepting proposals for this solicitation.

However, proposers may still submit written questions about solicitation topics through the DoD's SBIR/STTR Interactive Topic Information System (SITIS), in which the questioner and respondent remain anonymous and all questions and answers are posted electronically for general viewing until the solicitation closes. All proposers are advised to monitor SITIS (11.3 Q&A) during the solicitation period for questions and answers, and other significant information, relevant to the SBIR 11.3 topic under which they are proposing.

If you have general questions about DoD SBIR program, please contact the DoD SBIR Help Desk at (866) 724-7457 or email weblink.