You are here

Local Stochastic Prediction for UUV/USV Environmental Awareness

Award Information
Agency: Department of Defense
Branch: Navy
Contract: N68335-20-C-0567
Agency Tracking Number: N19A-022-0084
Amount: $1,996,697.00
Phase: Phase II
Program: STTR
Solicitation Topic Code: N19A-T022
Solicitation Number: 19.A
Solicitation Year: 2019
Award Year: 2020
Award Start Date (Proposal Award Date): 2020-07-28
Award End Date (Contract End Date): 2024-08-30
Small Business Information
11006 Clara Barton Dr.
Fairfax Station, VA 22039-1111
United States
DUNS: 116921678
HUBZone Owned: No
Woman Owned: No
Socially and Economically Disadvantaged: No
Principal Investigator
 Emanuel Coelho
 (228) 342-4773
Business Contact
 Kevin Heaney
Phone: (703) 346-3676
Research Institution
 Massachusetts Institute of Technology
 Pierre Lermusiaux
77 Massachusetts Avenue
cambridge, MA 02139-4301
United States

 (617) 324-5172
 Nonprofit College or University

This project delivers a compact system to assess and reduce local uncertainties that impact routing and sensor operation decisions while tracking the evolution of the maritime environment around unmanned platforms at sea (UUV/USV). The system runs both at control centers and on-board the UUV/USV’s, subject to different network bandwidth and computing environments Size, Weight and Power (SWaP) constraints. The system uses the Navy ocean forecasts for initial environmental guesses and outlooks for up to 2 weeks (or more in future generations) and then implements Reduced Order Models (ROM) to update the original forecast fields, along with a local uncertainty picture (for the next 24-48 hours). The ROM solutions target the variables and parameters of relevance for the UUV/USV fleet mission planning and execution (e.g. currents and sound speed). The reduced order estimates of the parameters and variables of interest are computed from a set of dynamic modes derived from ocean ensembles (e.g. perturbed using Gaussian Mixture-Models and updated through Dynamically Orthogonal functions or constrained by reduced physics solutions). The amplitudes of the reduced modes are updated at the control centers and sent to the UUV/USV platforms using small size signals (order KB) to enable reconstruction of the new local forecasts, using a pre-loaded reduced modes set. To ensure local fitness for short time-ranges, in-situ network observations are assimilated in-stride using machine learning solutions. The forecast reconstruction code and machine learning runs are executed on both reach-back centers and on dedicated payloads and used for path optimization and environmental adaptation/adaptive sampling.

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

US Flag An Official Website of the United States Government