You are here

Highly Integrated and Distributed Recognition Architecture for Anti-Ballistic Missile (HYDRA-ABM) System

Award Information
Agency: Department of Defense
Branch: Missile Defense Agency
Contract: HQ0860-22-C-7073
Agency Tracking Number: B212-020-0390
Amount: $149,995.00
Phase: Phase I
Program: SBIR
Solicitation Topic Code: MDA21-020
Solicitation Number: 21.2
Timeline
Solicitation Year: 2021
Award Year: 2022
Award Start Date (Proposal Award Date): 2021-12-06
Award End Date (Contract End Date): 2022-06-05
Small Business Information
888 Easy Street
Simi Valley, CA 93065-1111
United States
DUNS: 611466855
HUBZone Owned: No
Woman Owned: Yes
Socially and Economically Disadvantaged: Yes
Principal Investigator
 Francisco Maldonado
 (805) 582-0582
 emelgarejo@americangnc.com
Business Contact
 Emily Melgarejo
Phone: (805) 582-0582
Email: emelgarejo@americangnc.com
Research Institution
N/A
Abstract

American GNC Corporation (AGNC) and the University at Albany - State University of New York (UAlbany) propose the Highly Integrated and Distributed Recognition Architecture for Anti-Ballistic Missile (HYDRA-ABM) System. This technology is designed considering: (i) Missile Defense Systems requirements; (ii) a variety of networked sensors; and (iii) networked communications. High level system components are: (1) set of nodes, (2) parameter server, and (3) communication channels. The system design considers the effects of communication channel constraints related with the (i) sensor system communication module and (ii) characteristics of the networked communication to interconnect the set of nodes. The core benefit is a flexible and distributed object classification system optimized by Federated Learning and designed for networked, multimodal sensors with non-IID data. Specific innovations are: (1) a classification system with performance optimization by Federated Learning where the classification knowledge of each node is incorporated in the global learning process (and shared among nodes) while training data requirements are relaxed; (2) system architecture and software framework for integration of a variety of disparate missile defense sensors considering networked and multimodal sensors and data communication module and channels constraints; and (3) a novel system implementation for missile defense systems with global optimization using an advanced distributed learning process specifically tailored to meet DoD needs. Approved for Public Release | 21-MDA-11013 (19 Nov 21)

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government