An Objectively Optimized Observing System - PowerPoint PPT Presentation

1 / 36
About This Presentation
Title:

An Objectively Optimized Observing System

Description:

An Objectively Optimized Observing System – PowerPoint PPT presentation

Number of Views:19
Avg rating:3.0/5.0
Slides: 37
Provided by: crew
Category:

less

Transcript and Presenter's Notes

Title: An Objectively Optimized Observing System


1
An Objectively Optimized Observing System
  • David John Lary
  • NASA GSFC/UMBC GEST

2
Overview
  • Automated validation, bias detection, and cross
    calibration.
  • Objectively optimized observation direction for a
    sensor network.
  • Applications.

3
Automating Validation, Bias-Detection and
Cross-Calibration
4
Geophysical insights
c.a. 3000 B.C. Who can understand his errors?
Psalms 1912
5
Aura validation
6
Validation using PDFs
7
How shall we correct these biases?Can we
correct them on the fly?
8
Neurological algorithms
9
An example neural network
10
An example neural network
11
Instrument Inter-calibration
12
http//www.ndvicentral.info/
13
(No Transcript)
14
Moving towards autonomy
Goal Observing the right place at the right time!
15
Purpose
Wisdom is profitable to direct!
Wisdom is profitable to direct!
  • Scientific discovery through Inter-disciplinary
    research, observation and a high level of
    automation in the generation of state of the art
    models, visualization and data mining to
    facilitate understanding.
  • So far this work has won five NASA awards, a
    Royal Society Fellowship, an Alon Fellowship, a
    NASA Distinguished Goddard Fellowship in Earth
    Science and over 2.5 million dollars of funding.

16
Benefit Statement
  • Allows a focus on the key issue(s) whether they
    are
  • The societal needs/policy decisions
  • The decision support criteria
  • The scientific goals

17
Integrated Earth Observation
  • As we move to more comprehensive observing
    systems there will be more need for reusable
    intelligent observation direction systems.
  • These systems will respond in real time to the
    observation requirements.

18
Integrated Earth Observation
  • The requirements will be varied depending on the
    application.
  • The observing system will contain many
    components, orbital and suborbital.

19
Sensor WebA dynamic network of sensors working
to achieve a common goal.
Mosaic of Anacapa Island obtained using the DCS
camera during the Altair Channel Islands flight
The Ultra Long Duration Balloon (ULDB) is a
pumpkin-shaped balloon designed for durations
up to 100 days.
  • We are moving towards much more distributed
    observation systems with constellations of
    satellites and suborbital platforms, including
    autonomous vehicles.
  • There is a need for an intelligent observation
    direction system.

Current Altair payload
Altair information courtesy of Jim Elkins
http//uas.noaa.gov/altair/index.html
20
Observation scheduling
  • With flexible pointing instruments
  • What is the optimum real time pointing?
  • With flexible mode instruments
  • What is the optimum real time use of zoom in
    mode?
  • When should balloons be launched?
  • What is the optimum trajectories for UAVs and
    aircraft?

21
How do we achieve this?What do we optimize for?
22
The knowledge of ignorance is the beginning of
knowledge
The knowledge of ignorance is the beginning of
knowledge
  • The system envisioned will dynamically adapt the
    what, where, when and mode (e.g. global survey or
    zoom-in) of the observations made in an online
    fashion to maximize information content and
    minimize uncertainty in characterizing the
    systems state vector for a given observation
    capability.

23
Autonomous Observing Systems
  • Put simply, we use measures of what we do not
    know to direct our future observations, and use
    measures of how important it is to know it to
    determine the scheduling priority of
    observations.
  • What we do not know is quantified by state vector
    uncertainty supplied by the assimilation system.
  • How important it is to know it is quantified by
    information content also supplied by the
    assimilation system.
  • The geographic extent of the uncertainty maxima
    is one metric that can be used to determine
    whether zoom in or survey mode is required.

24
Geophysical insights
25
Chemical Data Assimilation
Uncertainty information is included
26
(No Transcript)
27
Key Features
Won 5 NASA awards
  • Rapidly reconfigurable
  • Automatic Code generation
  • Self documenting
  • Functionality
  • Results (automatic web site creation).

28
Interface
  • STK interface that can be used for prototyping or
    real time control.
  • Automatic web site creation for online
    data-mining and presentation.
  • Accessible for specialists and policy makers.

29
Hydrocarbon oxidation
30
Hydrocarbon oxidation
31
Health ApplicationsSimulation courtesy of Jay
Boris NRL
32
Logan Airport
33
Decision Support Systems
  • Using
  • NASA earth observing data (in this case
    including OMI HCHO)
  • EPA observing network
  • Navy Research Laboratory street level model
  • AutoChem chemistry scheme

34
(No Transcript)
35
Street level modeling
36
Wisdom is profitable to direct! (Solomon 3000
bp) The knowledge of ignorance is the beginning
of knowledge.
Write a Comment
User Comments (0)
About PowerShow.com