You are here


Award Information
Agency: National Aeronautics and Space Administration
Branch: N/A
Contract: 80NSSC21C0118
Agency Tracking Number: 212457
Amount: $124,982.00
Phase: Phase I
Program: STTR
Solicitation Topic Code: T10
Solicitation Number: STTR_21_P1
Solicitation Year: 2021
Award Year: 2021
Award Start Date (Proposal Award Date): 2021-05-03
Award End Date (Contract End Date): 2022-06-19
Small Business Information
7852 Walker Drive
Greenbelt, MD 20770-3208
United States
DUNS: 110592016
HUBZone Owned: No
Woman Owned: No
Socially and Economically Disadvantaged: No
Principal Investigator
 Kenneth Center
 (240) 391-3310
Business Contact
 Kenneth Center
Phone: (240) 391-3310
Research Institution
 University of Maryland
7809 Regents Drive, Lee Building 3112
College Park, MD 20742-5141
United States

 Nonprofit College or University

Orbit Logic (OL) and the University of Maryland College Park (UMd) are teaming to develop a machine learning solution, Lunar FLAPPER (Fault Learning Agent for Prediction, Protection and Early Response), that is able to operate in two modes: 1) a fully autonomous operations monitoring mode with fault detection, correction and reporting, and 2) a semi-autonomous mode, which can request approval for corrective actions and which can utilize human (ground operator) inputs to learn ideal operating conditions through positive re-enforcements of solution selections. OL will be leveraging and extending the work of our original FLAPPER SBIR research initiative to include operator in-the-loop training. We will enhance our already mature Autonomous Planning System (APS) product by adding Specialized Autonomous Planning Agents (SAPAs) to accommodate these new use cases.The end-vision includes both ground- and space-resident software components and tools. Lunar FLAPPER infrastructure on the ground will integrate with emulation, simulation, and real hardware instances to interact with mission data sources (telemetry and commands) - then observe data activity when mission operational scenarios are executed. Machine Learning (ML) techniques (already prototyped in previous NASA-funded efforts) will be utilized to learn what constitutes both nominal and off-nominal system behaviors, building both a library of ldquo;detection kernelsrdquo; (usable to recognize and characterize anomalous behavior) and a library of appropriate responses to correct faults and restore nominal operations. The goal is an eventual transition to as much onboard autonomy and fault management as possible, such that the station and all of its onboard systems are fully sustained with minimal or no intervention from Earth-based operators when it is not inhabited.

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

US Flag An Official Website of the United States Government