You are here

Federated Intelligence, Surveillance, Reconnaissance (ISR) Collection Management Using Machine Learning (ML)

Description:

TECHNOLOGY AREA(S): Information Systems

OBJECTIVE: The desired end product of Phase II is to have a federated collection management software that provides a coordinated collection plans across National to Tactical ISR collection to include Joint and Mission Partner Environments (MPE) using machine learning to optimize collection plans across the enterprise.

DESCRIPTION: The Army technical problem can be broken down into several areas as it relates to Multi-Domain Operations (MDO). First, current collection plan generation is performed in a silo approach based on mission objectives. Often times it is completed through spreadsheets and PowerPoint. Second, these collection plans are not visible or sharable to entities outside of unit organizations that create them. Third, this leads to redundant collection plans often times relying on ISR collection assets that would be collecting on plans that may have common objectives. This leads to inefficiencies and increased timeliness of critical information. Lastly, collection plans are largely manually generated which requires multiple human generated steps to develop an optimized collection plan that has no relationship to other collection plans that may have similar objectives or National/Tactical ISR collection assets could collect in route if multiple desperate collection plans are geo-spatially and temporally close.

PHASE I: Provide a concept that addresses the challenges related to objectives of this SBIR. As part of the concept define what the minimum viability of the capability will provide with the goal of increased functionality while providing a fluid user experience in Phase II. Phase I shall also address concepts for built in training and reminders for users to quickly operate and maintain proficiency of the system.

PHASE II: Provide a physical proof of concept system that showcases how a federated collection plan is implemented with some level of automation applied in the creation of collection plans. Additionally, the proof of concept system shall implement machine learning to optimize the federated collection plans while also enabling unit organizations to significantly improve their abilities to leverage existing or previously generated collection plans that potentially have similar objectives.

PHASE III: This SBIR would enable current Synchronized High Op-tempo Targeting (SHOT) S&T efforts that are part of the Long Range Precision Fires (LRPF) portfolio. SHOT does not currently have a federated ISR collection task. The SBIR would also enable the Tactical Intelligence Targeting Access Node (TITAN) program under PM DCGS-A which would provide a path for transition into an operational capability. Lastly, the commercial applicability of this effort shall support commercial geospatial capability providers as well as consumers that utilize commercial satellites, manned/unmanned aircraft for land/surface surveilling to include law enforcement.

KEYWORDS: ISR, MDO, MPE, National, Tactical, Collection Plans, Collection Management, SHOT, LRPF, Machine Learning, TITAN

References:

https://www.tradoc.army.mil/Portals/14/Documents/MDO/TP525-3-1_30Nov2018.pdf; https://apps.dtic.mil/dtic/tr/fulltext/u2/1051723.pdf; https://armypubs.army.mil/epubs/DR_pubs/DR_a/pdf/atp2_01.epub; https://www.jcs.mil/Portals/36/Documents/Doctrine/pubs/jp2_0.pdf

US Flag An Official Website of the United States Government