DoD SBIR 23.3 BAA

Active
No
Status
Closed
Release Date
August 23rd, 2023
Open Date
September 20th, 2023
Due Date(s)
October 18th, 2023
Close Date
October 18th, 2023
Topic No.
AF233-D025

Topic

Improved Data Collection and Knowledge Graphing in the TAK Ecosystem

Agency

Department of DefenseN/A

Program

Type: SBIRPhase: BOTHYear: 2023

Summary

The Department of Defense (DoD) is seeking proposals for improved data collection and knowledge graphing in the Tactical Assault Kit (TAK) ecosystem. The objective of this topic is to demonstrate the capability to define, capture, organize, label, and reason over the data generated by end-user devices and servers in the TAK ecosystem for use in machine learning model development, re-training, fine-tuning, and federated learning. The technology should leverage general-purpose machine learning tools, Android sensor hubs, and semantic network/knowledge graphing tools to extend the TAK-ML framework. The Phase I award is not required, and the offeror should provide documentation demonstrating accomplishment of a "Phase I-type" effort in the Direct to Phase II proposal. Phase II objectives include the development of technologies to collect and harness data from the TAK ecosystem for machine learning tasks, integrating with AFRL toolkits. Successful Phase II technologies may be candidates for Phase III development and potential transition to the TAK ecosystem or other AFRL programs. The project duration and funding specifics are not provided in the document. For more information, refer to the DoD SBIR 23.3 BAA solicitation notice on grants.gov.

Description

OUSD (R&E) CRITICAL TECHNOLOGY AREA(S): Advanced Computing and Software;Directed Energy (DE)

 

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 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: The objective of this topic is to demonstrate a capability to define, capture, organize, label, and reason over the data that is generated by end-user devices and servers in the Tactical Assault Kit (TAK) ecosystem for use by machine learning model development, re-training, fine tuning, and federated learning of existing models, or consumption by AFRL ML tools.

 

DESCRIPTION: The TAK ecosystem currently has a wealth of sensor data, usage data, and analytics that is under-utilized for artificial intelligence/machine learning (AI/ML). Leverage general-purpose machine learning (ML) tools (StreamlinedML/MISTK, WingmanAI), Android sensor hubs (TAK-ML sensor framework, Foresight and Sensor Manager), and semantic network/knowledge graphing tools (KnowML, FuelAI) to extend the TAK-ML framework. Accept analytics back from any frameworks, models, or plugins developed for further refinement. Use of digital engineering tools to at a minimum define open application programming interfaces (APIs) and where applicable build reference implementations is preferred.

 

PHASE I: This topic is intended for technology proven ready to move directly into a Phase II. Therefore, a Phase I award is not required. The offeror is required to provide detail and documentation in the Direct to Phase II proposal which demonstrates accomplishment of a “Phase I-type” effort, including experience with extension, modification, or creation of enterprise machine learning life cycle management toolkits for knowledge graphic, data curation, and related machine learning tasks.

 

PHASE II: Phase II objectives include the development of technologies to collect, reason over, and harness data from the TAK ecosystem for use in machine learning tasks, demonstrating integrations with (and extensions of) AFRL toolkits such as TAK-ML, StreamlinedML/MISTK, and KnowML to apply broader Air Force machine learning development to the tactical edge.

 

PHASE III DUAL USE APPLICATIONS: Successful Phase II technology effort reaching suitable TRL (6-7) will be candidates for additional Phase III development, including potential for transition to the Tactical Assault Kit (TAK) ecosystem in partnership with the TAK Product Center (TPC) or to other AFRL programs developing next generation AI/ML capabilities.

 

REFERENCES:

https://github.com/raytheonbbn/tak-ml;
https://dl.acm.org/doi/abs/10.1109/MILCOM52596.2021.9652909;
https://mistkml.github.io/;
https://tak.gov;
https://civtak.org;

 

KEYWORDS: semantic web; knowledge graphing; mobile machine learning; end-user devices; data analytics; ATAK