Navy Technology Acceleration - Machine Learning (ML) and Artificial Intelligence (AI) to Develop Capabilities and Impact Mission Success
Navy SBIR 2019.3 - Topic N193-A01
Opens: September 24, 2019 - Closes: October 23, 2019 (8:00 PM ET)

N193-A01

TITLE: NAVY TECHNOLOGY ACCELERATION - Machine Learning (ML) and Artificial Intelligence (AI) to Develop Capabilities and Impact Mission Success

 

TECHNOLOGY AREA(S): Information Systems

ACQUISITION PROGRAM: Broad ranging topic related to AI/ML in support of the Navy Technology Acceleration Pilot.

The technology within this topic is restricted under the International Traffic in Arms Regulation (ITAR), 22 CFR Parts 120-130, which controls the export and import of defense-related material and services, including export of sensitive technical data, or the Export Administration Regulation (EAR), 15 CFR Parts 730-774, which controls dual use items. Offerors must disclose any proposed use of foreign nationals (FNs), their country(ies) of origin, the type of visa or work permit possessed, and the statement of work (SOW) tasks intended for accomplishment by the FN(s) in accordance with section 3.5 of the Announcement. Offerors are advised foreign nationals proposed to perform on this topic may be restricted due to the technical data under US Export Control Laws.

OBJECTIVE: Develop artificial intelligence (AI)/machine learning (ML) capabilities to address a variety of use cases that expand outside the current field of focus of the Navy. Technologies should address capability development, testing and certifying AI/ML algorithms, Readiness and Sustainment, as well as enable analyses of massive quantities of data in a multitude of applications with a shared focus on program and fleet success.

DESCRIPTION: The Department of the Navy is interested in the development of cutting-edge AI/ML technologies and intends to collaborate with innovative small businesses to obtain solutions to the following and related Navy Focus Areas. Submit no more than one proposal per topic to one of the following Focus Areas: 

1 - Readiness and Sustainment
2 - Unmanned Aircraft Systems Autonomy and Automation
3 - Predictive Maintenance
4 - Cyber
5 - Counter Artificial Intelligence
6 - Streamline Business Operations
7 - Integration of Automatic Dependent Surveillance
8 - Integration of Automatic Identification System (AIS) Data through AI/ML Applications
9 - C4ISR (Test/Certify)

1. Readiness and Sustainment - Maintaining inventories and supply chains is a critical function within Naval Air operations; this becomes especially important in keeping a frontline offensive supplied and ready. This process involves keeping suppliers aware of current demands and the flow of supplies to the destination. Aircraft readiness depends significantly on efficient supply chain. Currently, the acquisition software and databases embedded with bad data make it difficult to track parts. This affects the prediction of supply chain needs, making detection by humans improbable. Errors in the data propagate within the databases, causing major delays. Using AI/ML protocols to identify such errors, and applying deep learning techniques with pattern analysis, can cleanse the data error in short intervals. AI/ML protocols can also uncover relationships between variables and clusters, currently an expertise limited to experts.

Develop innovative AI/ML technologies that can predict and prescribe items for resupply. Develop innovative technologies that utilize AI/ML techniques and collaborative planning to address efficient logistics support, maintain inventories, reduce waste, allocate spare parts, and optimize inventory levels. Demonstrate scalability and trouble-shooting to enable rapid deployment of agile, adaptable forces at reduced costs. Successful development will enable the warfighter to receive the correct material at the right time and place, contributing to increased readiness and sustainment.

2. Unmanned Aircraft Systems Autonomy and Automation - Develop AI/ML solutions for unmanned systems with a focus on capability development, autonomy, and automation. Software architectures and systems capabilities often define Navy unmanned assets whether they are unmanned aerial systems (UAS) or weapon systems. Accurate perception of the surroundings is critical to accomplish unmanned missions. Work in the area of image understanding of "standard" electro-optical/visual (EO) imagery has been characterized by sharp, well lit, and well framed features, rather than lesser quality images or "non-optical" imagery such as those from IR (infrared), SAR (synthetic aperture radar), and ISAR (inverted SAR).

Explore and develop advanced image understanding techniques, such as multimodal imagery, in conjunction with sensor fusing. Architectures and implementations contain vulnerabilities that put survivability of systems at risk, often making them the target of cyber-attacks. Leverage AI/ML techniques to design, develop, and test processes that increase the resilience and survivability of critical UAS/weapon/weapons systems software through optimization of implementation and architectures that consider both failures due to mistakes and events perpetrated by adversaries.

3. Predictive Maintenance - Predictive maintenance applications, such as condition-based maintenance, have huge potential for supporting fleet forces and driving efficiencies. Develop novel approaches that predict and mitigate the failure of critical parts, target aircraft mission degraders such as foreign object debris and corrosion, automate diagnostics, and plan maintenance based on data and equipment conditions. Produce prototypes of predictive maintenance solutions and demonstrate scalability. Such AI/ML based applications have the potential to predict, more accurately, maintenance needs on equipment. Such solutions will significantly improve availability of aircraft on the flight line, increase operational readiness, and reduce life cycle costs.

4. Cyber - Cyber risk assessment and management of the Navy's weapons and weapons systems, quantification and understanding of risk provides temporary results based upon information available at the time of the assessment, and the risk to platforms in cyber-contested environments changes rapidly.  Develop tools and techniques using AI/ML and analytic techniques to accumulate and integrate internal/external information, to report risk in near real-time. The developed system should be able to identify trends and emerging risks based on historical and current information, as well as provide risk measures of the mission through the development of key risk indicators, key performance indicators, and associated threat measures. The resulting system would extend the concept of CYBERSAFE to a near real-time environment using the results of those processes as a baseline.

5. Counter Artificial Intelligence - Methods used to trick AI/ML techniques, something as simple as changing a pixel in a common picture derived out of AI/ML techniques, can lead to misclassification of the image, resulting in unintended consequences; the system programmed to identify the subject of the photo is unable to do that through a small tweak. That said we must to determine if AI/ML can be trusted to interpret data correctly and act accordingly.

Develop innovative approaches such as complimentary classifiers and meta-reasoners to understand such failure modes, propose mitigation plans to prevent deceit of AI/ML algorithms, leading to resilient systems. Such solutions enhance AI/ML techniques’ capabilities, delivering results that can be trusted and validated, and on par with human-like performance.

6. Streamline Business Operations – The DoD workforce dedicates time and effort on highly manual, repetitive tasks that are prone to errors. AI/ML technologies have the potential to reduce the number and cost of mistakes, increase productivity, and allow allocation of DoD resources to higher-level and mission-priority activities. As an example, the workforce is investing significant time and money to assess the current state of projects, with respect to cost, schedule, and performance. Often, the earned value management processes fall short of identifying real problems with a project during its duration. Data driven AI/ML techniques could identify such risks, optimize allocation of resources, and automate mundane project tasks. Develop innovative approaches applying AI/ML techniques for project management capacities, human capital management, workforce productivity and efficiency enhancement, and automation of business systems and digital workflow, which connect data and processes at the enterprise level to drive better business outcomes.

7. Integration of Automatic Dependent Surveillance – Broadcast (ADS-B) data through AI/ML Applications: The ADS-B data are obtained from publicly available sources.  The Navy seeks to develop models and algorithms through AI/ML processes to autonomously characterize behaviors of self-reporting aircraft using ADS-B data. The behavior models and data will be used to (1) identify apparent air corridors and (2) detect anomalous behavior in support of determining aircraft intent.

8. Integration of Automatic Identification System (AIS) Data through AI/ML Applications - AIS data are obtained from publicly available sources. The Navy seeks to develop models and algorithms using AI/ML processes to autonomously characterize behaviors of self-reporting maritime traffic using AIS data in order to use these behavioral models and data to (1) identify apparent shipping lanes and (2) detect anomalous behavior in support of determining surface vessel intent.

9. C4ISR (Test/Certify) – Trusted and reliable AI technologies can be used to enhance mission capability and increase the performance of many types of Naval systems. Recent advances in ML are improving countless technologies from image classifiers to game playing, with the potential to revolutionize innumerable others, from natural language processing to robotics. However, the current ability to leverage advancements are limited because no reliable method exists for testing and certification of the outputs of these systems. Therefore, the Navy is seeking innovative solutions to enable the transformation of opaque ML and AI systems into trusted and understandable systems, necessary for the warfighter to utilize these advanced systems reliably to achieve mission goals.

Develop appropriate framework and methods to test and certify ML and AI algorithms and systems using ML and AI technologies for Program Executive Office for Command, Control, Communications, Computers and Intelligence (PEO C4I). Successful methods will provide an effective and efficient way to test and certify Navy systems utilization of ML and AI algorithms and allow acquisition and fielding.

Awardees should conduct testing in an operationally relevant environment with final testing by the Navy. Validation, testing, qualification, and certification for Navy use across a wide range of conditions as applicable for the relevant class of problem will be conducted.

Work produced in Phase II may become classified. Note: The prospective contractor(s) must be U.S. Owned and Operated with no Foreign Influence as defined by DOD 5220.22-M, National Industrial Security Program Operating Manual, unless acceptable mitigating procedures can and have been be implemented and approved by the Defense Security Service (DSS). The selected contractor and/or subcontractor must be able to acquire and maintain a secret level facility and Personnel Security Clearances, in order to perform on advanced phases of this contract as set forth by DSS and the awarding NAVY SYSCOM in order to gain access to classified information pertaining to the national defense of the United States and its allies; this will be an inherent requirement. The selected company will be required to safeguard classified material IAW DoD 5220.22-M during the advance phases of this contract.

PHASE I: NOTE: Please add the Focus Area number you are proposing to as a prefix to the Phase I Proposal title.

Develop a solution to address one or more of the use cases outlined in the Description and demonstrate the feasibility of that concept. Assure data integrity that is representative of affected processes. Feasibility can also be established through modeling, simulation, and analysis. A high-level description of the intended approach for Phase II should be included in the Phase I proposal.

PHASE II: Based upon the results of Phase I, develop, demonstrate functionality and deliver prototype systems for testing and evaluation. The prototype system will vary based on the proposed approach, but it may include hardware and software.
 
It is probable that the work under this effort could become classified under Phase II (see Description section for details).

PHASE III DUAL USE APPLICATIONS: Transition the technology developed to improve and expand mission capability to a potentially broad range of government programs and entities. Commercialize the various technologies developed to civilian entities with alternate mission needs.

REFERENCES:

1. VADM Dean Peters article (USNI June 14, 2018) calling for readiness improvements that our AI application is uniquely qualified to enable throughout the FRCs depots space. https://news.usni.org/2018/06/14/navair-to-develop-modernization-plan-for-3-depots

2. During October 2018 NRDE A2I Summit in San Diego, RADM David Hahn challenged attendees (government and industry) to find ways to "take AI to scale" and to accelerate AI-enabled technologies into the Fleet "at the speed of industry."

3. “Summary of the 2018 department of defense artificial intelligence strategy”, Accessible from  https://media.defense.gov/2019/Feb/12/2002088963/-1/-1/1/SUMMARY-OF-DOD-AI-STRATEGY.PDF, February 2019.

4. U.S. Department of Homeland Security, “Automatic Identification System Overview”, United States Coast Guard. 17 November 2018 https://www.navcen.uscg.gov/?pageName=aismain.

5. Bishop, Christopher.  Pattern Recognition and Machine Learning.  New York, Springer-Verlag, 2006
https://www.springer.com/us/book/9780387310732.

6. U.S. Department of Transportation, “Automatic Dependent Surveillance-Broadcast (ADS-B)”, Federal Aviation Administration. 17 November 2018. https://www.faa.gov/nextgen/programs/adsb/.

7. Castelvecchi, D. The Black Box of AI. Nature 538, 2016, pp. 20-23.

8. Russell, S.R., and P. Norvig. Artificial Intelligence: A Modern Approach. Upper Saddle River, NJ: Prentice Hall, 2015.

KEYWORDS: Artificial Intelligence; Neural Networks; Big Data; Machine Learning (ML); Data Analysis; Sustainment and Readiness; Automatic Dependent Surveillance-Broadcast (ADS-B); Automatic Identification System (AIS); Testing & Evaluation; Certification

TPOC-1:

SBIR Navair Contact

Phone:

301-995-1825

 

TPOC-2:

Rebecca Gassler (Focus Areas 7 - 8)

Phone:

202-781-2751

Email:

rebecca.gassler@navy.mil

 

TPOC-3:

Onekki Montgomery (Focus Areas 7 - 8)

Phone:

202-781-1186

Email:

onekki.montgomery@navy.mil

 

TPOC-4:

Raphael Pei (Focus Area 9)

Phone:

619-524-4536

Email:

raphael.pei@navy.mil

 

** TOPIC NOTICE **

These Navy Topics are part of the overall DoD 2019.3 SBIR BAA. The DoD issued its 2019.3 BAA SBIR pre-release on August 23, 2019, which opens to receive proposals on September 24, 2019, and closes October 23, 2019 at 8:00 PM ET.

Direct Contact with Topic Authors. From August 23 to September 23, 2019 this BAA is issued for Pre-Release with the names of the topic authors and their phone numbers and e-mail addresses. During the pre- release period, proposing firms have an opportunity to contact topic authors by telephone or e-mail to ask technical questions about specific BAA topics. Questions should be limited to specific information related to improving the understanding of a particular topic’s requirements. Proposing firms may not ask for advice or guidance on solution approach and you may not submit additional material to the topic author. If information provided during an exchange with the topic author is deemed necessary for proposal preparation, that information will be made available to all parties through SITIS (SBIR/STTR Interactive Topic Information System). After this period questions must be asked through SITIS as described below.

 

SITIS Q&A System. Once DoD begins accepting proposals on September 24, 2019 no further direct contact between proposers and topic authors is allowed unless the Topic Author is responding to a question submitted during the Pre-release period. However, proposers may submit written questions through SITIS at https://sbir.defensebusiness.org/topics . In SITIS, the questioner and respondent remain anonymous and all questions and answers are posted electronically for general viewing.

Topics Search Engine: Visit the DoD Topic Search Tool at sbir.defensebusiness.org/topics/ to find topics by keyword across all DoD Components participating in this BAA.

Help: If you have general questions about DoD SBIR program, please contact the DoD SBIR Help Desk at 800-348-0787 or via email at sbirhelpdesk@u.group