DON26BZ05-DV087 TITLE: DIRECT TO PHASE II: Edge-Deployed Explainable Digital-Twin Condition Based Maintenance CBM+ Platforms for Carrier-Based Systems
OUSW (R&E) CRITICAL TECHNOLOGY AREA(S): Applied Artificial Intelligence (AAI)
COMPONENT TECHNOLOGY PRIORITY AREA(S): Advanced Computing and Software
PROJECTED CMMC LEVEL REQUIREMENT: Level 2 (Self)
OBJECTIVE: Develop Digital Twin modeling and simulation, leveraging existing and new sensors to detect faults and predict possible failures in Aircraft Launch and Recovery Equipment (ALRE) to provide advance notice and prepare "maintainers" performing Condition Based Maintenance (CBM).
DESCRIPTION: Carrier-based aviation platforms under NAVAIR generate massive, high-frequency streams of voltage, current, pressure, vibration, and temperature data across multiple subsystems—from the Advanced Arresting Gear (AAG) and steam catapults to hydraulic deck handlers and fuel-management valves—but Denied, Disrupted, Intermittent, Limited (DDIL) connectivity in carrier deployment forces maintenance crews to batch-download raw logs via physical media and ferry them ashore for analysis, introducing critical delays that can obscure early indicators of seal leakage, actuator fatigue, or control-valve drift, drive up unscheduled maintenance and lifecycle costs, and jeopardize sortie rates and aircrew safety in forward-deployed environments. DoDI 4151.22 mandates that CBM+ be "examined, evaluated, integrated, and incorporated" into weapon-system engineering and sustainment plans to optimize readiness and reduce life-cycle costs, and OPNAVINST 4790.16C requires Navy program managers to document CBM+ implementation and assess CBM+ maturity at every acquisition and sustainment review. By embedding a reusable, edge-deployed digital-twin core with semantic-AI reasoning, CBM+ workstations directly fulfill these requirements, enabling on-platform prognostics, formal CBM+ documentation, and continuous maturity assessments throughout NAVAIR system development and fielding.
Topic Focus: The Navy seeks proposals for an edge-deployed (at the point of need, e.g., on the ship), explainable (i.e., Explainable Artificial Intelligence) Digital-Twin CBM+ Platform—hereafter "CBM+ Workstation"—that:
1. Ingests & Reduces High-Rate Sensor Streams Onboard. Leverage model-order-reduction and adaptive sampling to transform raw sensor traces (e.g., water-twister pressure, hydraulic fluid temperature, vibrations, noise, electric-motor power conditioning system sensors, etc.) into compact, information-rich features in real time into compact, high-value analysis that facilitates transmission under DDIL limitations.
2. Embeds Hybrid Physics & Semantic-AI Reasoning. Fuse first-principles Digital Twin models with maintenance-domain knowledge graphs and shipboard asset and contextual mission-specific knowledge graphs to automatically infer and explain emerging fault modes, delivering human-readable diagnostics rather than opaque alerts, and providing fault isolation to the Lowest Replaceable Unit (LRU).
3. Supports Plug-and-Play Workstation Generation. Provide a secure developer API to instantiate customer-specific CBM+ "workstations" for any carrier-based subsystem, enabling rapid customization and field upgrades without source-code changes.
4. Meets Shipboard Resilience Standards. Demonstrate performance on MIL-SPEC or ruggedized hardware, with <1 second anomaly-detection latency and <10 MB/hour data-transfer footprints to shore when intermittent links reconnect.
5. Provide integration via high-level APIs with shore-based analytics and logistics processes.
By funding this topic, the Navy will catalyze a reusable, standards-based CBM+ solution family—accelerating deployment of predictive-maintenance workstations across the fleet, reducing unscheduled downtime, and ensuring mission readiness even in contested, communication-limited environments.
PHASE I: For a Direct to Phase II topic, the Government expects that the small business will have accomplished the following in a Phase I-type effort and developed a concept for a workable prototype or design to address, at a minimum, the basic requirements of the stated objective above. The following actions would be required to satisfy the requirements of Phase I:
Established the feasibility of the edge-deployed, explainable CBM+ Workstation by focusing on a representative NAVAIR subsystem, such as the Advanced Arresting Gear (AAG).
FEASIBILITY DOCUMENTATION: Offerors interested in participating in Direct to Phase II must include in their response to this topic Phase I feasibility documentation that substantiates the scientific and technical merit and Phase I feasibility described in Phase I above has been met (i.e., the small business must have performed Phase I-type research and development related to the topic NOT solely based on work performed under prior or ongoing federally funded SBIR/STTR work) and describe the potential commercialization applications. The documentation provided must validate that the proposer has completed development of technology as stated in Phase I above.
PHASE II: Develop a prototype CBM+ workstation core that operates on representative edge-computer hardware and processes live data from an AAG test site. Demonstrate semantic-AI fault inference for at least two critical AAG failure modes (such as seal leakage or valve blockage), generating confidence metrics and natural-language explanations to decrease maintenance downtime and improve AAG availability. Utilize existing sensor systems to perform electrical and mechanical diagnostics and prognostics. Create a software interface capable of automatically generating a lightweight CBM+ workstation for a second carrier subsystem, like the catapult accumulator system.
PHASE III DUAL USE APPLICATIONS: Transition the CBM+ Workstation from a prototype to a fully operational, supported, and scalable solution for the MK15 Advanced Arresting Gear and the Navy, while also pursuing broader commercial applications.
Commercial or industrial sites with limited network connectivity could greatly benefit from an edge-deployed digital twin for a variety of systems (additive manufacturing, HVAC systems, etc.).
REFERENCES:
KEYWORDS: Diagnostics; Prognostics; Health management; Digital twin; CBM+; Edge-compute
| 9/4/26 | Q. | The Phase II description states that the prototype CBM+ workstation core will process live data from an AAG test site, and a posted Q&A response states that the system must be designed to eventually be integrated on the MK15 Advanced Arresting Gear. Is the Phase II demonstration expected to remain in the test-site or laboratory environment for the duration of the base period of performance, or does the Government anticipate a shipboard installation, with the attendant cybersecurity authorization activity, within the Phase II base period? |
| A. | The government does not expect a shipboard transition during the Phase II contract. Phase II will be a combination of utilizing existing data that is provided to the contractor, and potential testing at a CONUS test site such as Jet Car Track Site (JCTS) track 4, where there is a representative example of AAG. The goal of this effort is to have a technology that will eventually be transitioned to the ship for long-term use, but that is outside the scope of a Phase II contract. Such a transition (assuming the Phase II is successful) is expected to be a part of a PMA-funded Engineering Change Proposal (ECP). | |
| 8/26/26 | Q. | 1. The FEASIBILITY DOCUMENTATION section requires that the small business has performed Phase I-type research and development related to the topic, not solely based on work performed under prior or ongoing federally funded SBIR/STTR work, and the Phase I description frames feasibility around a representative NAVAIR subsystem such as the Advanced Arresting Gear. Would self-funded commercial prior work performed outside naval aviation, which established an edge-deployed, explainable condition-analysis capability ingesting high-rate sensor and log streams and producing findings that remain traceable to the underlying source records, satisfy the feasibility documentation requirement, provided the proposal includes a credible technical mapping from that work to a representative ALRE subsystem for the Phase II effort?
2. Of the five capabilities listed under Topic Focus, should offerors treat the plug-and-play workstation generation capability (the secure developer API for instantiating customer-specific CBM+ workstations for additional carrier-based subsystems) as the primary innovation sought, or are the physics-based Digital Twin models the evaluation priority for Phase II? 3. What existing CBM+ tools, processes, or systems of record does the Government consider the current baseline for ALRE maintenance, which offerors should study and aim to improve upon? |
| A. | 1. Yes. The ultimate goal of this topic is to serve AAG, but we cannot expect a company to have literally demonstrated it used on AAG at the proposal phase. We expect a demonstrated proof of concept on something comparable, ex an electric motor and/or hydraulic systems.
2. We leave it up to the small business; topics are intentionally written vague so as not to pigeonhole the small business into any one particular focus. We simply have a problem (the need to diagnose and prognosticate issues) and are interested in technologies to achieve that in an “edge” system that is not dependent on connectivity (which may not be available). 3. There exists several sensors on AAG, primarily on the electric motor, and a few on the water twister (pressure and temperature). The software for this currently is binary: is it safe to take an arrestment or not. If this data can be used, or if new sensors are proposed, both options are acceptable for this topic. |
|
| 8/25/26 | Q. | 1. How many awards do you expect to make based on this topic?
2. What representative data, artifacts, models, or mission traces will be available to performers for the enterprise engineering math and deployment automation prototype? 3. What government-furnished interfaces, environments, or software versions must the prototype support in Phase I and Phase II? 4. What evaluation metrics will matter most at selection: accuracy, latency, operator workload, explainability, integration effort, or transition readiness? 5. Will performers be able to use unclassified or synthetic data for the Phase I feasibility demonstration? 6. Are there required hardware, sensor, platform, or edge-compute constraints that should be assumed before proposal submission? 7. Who is the intended transition owner or operational end user, and can proposers cite that office as the target customer? |
| A. | 1. This is a direct-to-Phase-II, so we anticipate just one award for this topic.
2. If you are selected for a contract, we will provide you schematic drawings of the system, as well as existing data, for use in the development of your technology. 3. There is only Phase II, but it must be designed to eventually be integrated on the MK15 Advanced Arresting Gear (AAG). 4. We will select the technology that is most likely to transition at the conclusion of Phase-II, where the program office (PMA-251) will be willing to fund its implementation on Ford-class aircraft carriers. The PMA is aware and endorsed this topic, but they have final say as to if it is transitioned later on. 5. Any demonstration (subscale or modeling / simulation) of synthetic data that is comparable to AAG will suffice. 6. The only real hard requirement is that this system can operate at the edge, where it is entirely stand-alone and not dependent on a data link or cloud server or anything like that to function. If the ship is in a conflict, communication will be limited or even non-existent. If updates are submitted when the ship is at home (ex every 6-12 months), that is fine. 7. The ultimate transition owner is PMA-251. Ideally, at the conclusion of this effort, PMA-251 will be satisfied such that they fund this technology being implemented on Gerald Ford CVN-78 class aircraft carriers. |
|
| 8/21/26 | Q. | 1. AAG data availability -
Will the Government provide historical and/or live AAG sensor data during Phase II, including labeled examples of known failure events?
2. Sensor interfaces - What existing AAG sensor interfaces, protocols, sampling frequencies, and data formats should offerors assume? 3. Failure-mode priorities - Beyond seal leakage and valve blockage, does the Government have a prioritized list of AAG failure modes that should drive Phase II diagnostics and prognostics? 4. Ground truth - What maintenance, inspection, or teardown information will be available to establish ground truth for fault detection and LRU isolation? 5. Digital-twin fidelity - What level of physics-based modeling does the Government expect—component-level first-principles simulation, reduced-order models, system-level state estimation, or another level? 6. Prognostics definition - Does "predict possible failures" require remaining-useful-life estimates, probability of failure within a time horizon, degradation trend prediction, or primarily advance fault detection? 7. Explainability - Are there specific human-machine-interface or explainability requirements for maintainers beyond confidence metrics and natural-language explanations? 8. LRU requirement - Should all demonstrated failure modes be isolated to an LRU, and what constitutes acceptable LRU-isolation accuracy? |
| A. | (1) If one is selected for a contract, that vendor will receive historical examples of AAG sensor data during Phase II, as well as drawings and a site visit to see the relevant equipment in-person.
(2) There exist various sensors on the electric motor (currents and voltages), as well as a few sensors on the water twister (pressure and temperature); these are used for health-map real-time monitoring. Currently, they only provide a Red-Deck / Green-Deck alert, with no diagnostics or prognostics. (3) There are no specific failure rates, but there is concern of fatigue, fracture, bearing failure, as well as electrical burnouts. (4) There is limited such data availability, though it is realistic to plan for an inspection process at the test site. At Lakehurst, there is a representative example of AAG that exists for testing. (5) The government does not have any specific requirement for physics-based modeling, AI, machine-learning, etc. There simply is the acknowledgement that there will be limited sensor data, and that any solution will obviously involve some form of analysis to extrapolate a developing fault before it happens. (6) All of those will be useful, but the primary goal is advanced fault detection, and really to drive the maintenance, to alert the sailors of maintenance needs, and avoid needless redundant or excessive maintenance. (7) There are no specific requirements, provided a typical enlisted sailor can comprehend the need to do inspections, maintenance, or equipment replacement in-situ on an aircraft carrier, with no contact to the shore (ie Edge Computing). (8) There is no specific requirement between maintenance and replacing an item. Our goals are to alert the sailors of impending problems, and guide them to required inspections or maintenance. |
** TOPIC NOTICE ** |
The Navy Topic above is an "unofficial" copy from the Navy Topics in the DoW FY-26 Release 5 SBIR BAA. Please see the official DoW Topic website at www.dodsbirsttr.mil/submissions/solicitation-documents/active-solicitations for any updates. The DoW issued its Navy FY-26 Release 5 SBIR Topics pre-release on August 5, 2026 which opens to receive proposals on August 26, 2026, and closes September 23, 2026 (12:00pm ET). Direct Contact with Topic Authors: During the pre-release period (August 5, through August 25, 2026) proposing firms have an opportunity to directly contact the Technical Point of Contact (TPOC) to ask technical questions about the specific BAA topic. The TPOC contact information is listed in each topic description. Once DoW begins accepting proposals on August 26, 2026 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. DoD On-line Q&A System: After the pre-release period, until September 9, 2026, at 12:00 PM ET, proposers may submit written questions through the DoW On-line Topic Q&A at https://www.dodsbirsttr.mil/submissions/login/ by logging in and following instructions. In the Topic Q&A system, the questioner and respondent remain anonymous but all questions and answers are posted for general viewing.
DoW Topics Search Tool: Visit the DoW Topic Search Tool at www.dodsbirsttr.mil/topics-app/ to find topics by keyword across all DoW Components participating in this BAA.
|