Software ProgramsPrograms - WordPress.comThe COMSOL® enviro li i h i l i t nm ent empowers you to co l i i l i h i vr e nduct d li capabilities.CAD-impo rea st c tec n ca s mu interfaces

September 2008

Special Supplement to

Sponsored by

Co-sponsored by

Inside: 45 New Software Programs

Inside: 45 New Software Programs

Cov ToC + – ➭

➮

AIntro

http://www.abpi.net/ntbpdfclicks/l.php?200809STBCVR1A

http://www.abpi.net/ntbpdfclicks/l.php?200809STBCVR1B

http://www.abpi.net/ntbpdfclicks/l.php?200809STBCVR1C

www.comsol.comCOMSOL, COMSOL MULTIPHYSICS, COMSOL SCRIPT AND COMSOL REACTION ENGINEERING LAB ARE REGISTERED TRADEMARKS OF COMSOL AB.

OTHER PRODUCT OR BRAND NAMES ARE TRADEMARKS OR REGISTERED TRADEMARKS OF THEIR RESPECTIVE HOLDERS. © 2008 COMSOL, INC. ALL RIGHTS RESERVED.

The COMSOL® environment empowers you to conduct

realistic technical simulations using multiphysics modeling

capabilities. CAD-import and application-specific user

interfaces make problem set-up easy. COMSOL’s ability to

model any coupled physics, along with its multicore parallel

solvers, facilitates best-in-class flexibility, accuracy and speed

to your simulations.

The Unifying Multiphysics Simulation Environment!

TT

The UnifThe Uniff

SimuuSim

fying Multffying Multulaation Enulation En

ysicstiphysicstiphnviroonmevironmen

ent!ent!

SimuuSim

u

LOSMOCehT ® orivne

ilihiil

lattion Enulation En

n

ocotuoysreowpmetnemno

ihiliil

virronmevironmen

e

t cudno

ild

nt!ent!

opmi-DAC.seitilibaapc

muislacinhcetcitsilaer

elborpekamsecafretni

yhpdelpuocynaledom

tsebsetatilicaf,srevlos

.snoitaalmuisruoyot

ucfiiceps-noitaacilpapdnatro

mscisyhpitlmugnisusnoitalu

as’LOSMOC.yy.saepu-tesme

rocitlmustihtiwgnola,scisy

ycarucca,yy,tilibixeflssalc-ni-t

r esu

g niledom

otytiliba

lellaraper

deepsdna

.cwwww.

RBROTCUDORPREHTO

TLUMLOSMOC,LOSMOC

comsol.com

TFOSKRAMEDARTDERETSIGERROSKRAMEDARTERASEMANDNAR

NIREENIGNENOITCAERLOSMOCDNATPIRCSLOSMOC,SCISYHPIT

SERSTHGIRLLA.CNI,LOSMOC8002©.SREDLOHEVITCEPSERRIEHT

.BALOSMOCFOSKRAMEDARTDERETSIGERERABALG

.DEVRES

Cov ToC + – ➭

➮

AIntro

http://www.abpi.net/ntbpdfclicks/l.php?200809STBCVR2

Free Info at http://info.hotims.com/1 38-790

Go to comsol.com/intro for a free multiphysics intro kit>

Learn more!

SIGNALS & SYSTEMS LAB

OPTIMIZATION LAB

COMSOL REACTION ENGINEERING LAB ®

RF MODULE

EARTH SCIENCE MODULE

MEMS MODULE

STRUCTURAL MECHANICS MODULE

MATERIAL LIBRARY

ACOUSTICS MODULE

HEAT TRANSFERMODULE

CHEMICAL ENGINEERING MODULE

AC/DCMODULE

CAD IMPORT MODULE

COMSOL SCRIPT ®COMSOL MULTIPHYSICS ®

EXAMPLE MULTIPHYSICS APPLICATIONS

• Electronics cooling

• Thermal processing

• Acoustics and vibrations in fluids and solids

• Bioengineering

• Electromechanical machinery

• Fluid-structure interactions

• Thermo-structural effects in antennas and waveguides

• Microfluidics

• Polymer mechanics

• Resistive and inductive heating

• MEMS

• CFD

ELPMAXE

ITAATCILPPASCISYHPITLUM

SNO

EELLPPMMAAXXEE

luid-structF•

omeclectrE•

ioengineerB•

coustics aA•

hermal prT•

onics lectrE•

IITTAAATATCCIILLPPPPAASSCCIISSYYHHPPIITTLLUUMM

e interactionstur

yhanical machiner

ring

and vibrations in fluids and solids

ocessing

cooling

SSNNOO

FDC•

MSEM•

e heatinge and inductivesistivR•

ymer mechanicsolP•

ofluidicsMicr•

ects in anhermo-structural effT•

g

eguidesvntennas and wa

CHEMICAL ENGINEERMODULE

AC/DCMODULE

RFMODULE

EARTH SCIENCERTRTMODULE

RING

SIGNALS &SYSTEMS LAB

O

OPTIMIZ TTION LABAATAT

COMSOL REACTIONENGINEERING LAB ®

MATERIAL LIBRARAATAT

ACOUSTICSMODULE

HEAT TRANSFERATATMODULE

MEMS MODULE

STRUCTURAL MECHANICSMODULE

YYRRYRY

R

CAD IMPORTRTRTMODULE

o to G comsultiphee mfr>

omnraeLS

osol.com/intr or a fo kitysics intrh

!ero

51

Cov ToC + – ➭

➮

AIntro

http://www.abpi.net/ntbpdfclicks/l.php?200809STBP1A

http://www.abpi.net/ntbpdfclicks/l.php?200809STBP1B

2 www.techbriefs.com Software Tech Briefs, September 2008

September 2008

Supplement to NASA Tech Briefs’ September 2008 Issue. Published by Tech Briefs Media Group

D E S I G N & A N A LY S I S S O F T W A R E

4 Multiphysics Modeling Simplifies Analysis of Electricaland Electromagnetic Effects

7 Geophysical Wave Propagation Calculation UsingMultiphysics

8 Simulating the Manufacturing Process of Ceramic MatrixComposites

9 3D Program Aids Design of Custom Iron Work

9 Displaying CFD Solution Parameters on Arbitrary CutPlanes

10 Flow Solver for Incompressible 2-D Drive Cavity

10 Flow Solver for Incompressible Rectangular Domains

E L E C T R O N I C S / C O M P U T E R S

10 Simulating Avionics Upgrades to the Space Shuttles

11 Simulating the Phoenix Landing Radar System

11 Injecting Artificial Memory Errors Into a RunningComputer Program

12 Fault-Tolerant, Multiple-Zone Temperature Control

12 Implementing a Digital Phasemeter in an FPGA

M E C H A N I C S / M A C H I N E R Y

13 Post-Flight Estimation of Motion of Space Structures: Part 1

13 Post-Flight Estimation of Motion of Space Structures: Part 2

13 Simulating Operation of a Large Turbofan Engine

14 Automated Assistance for Designing Active MagneticBearings

14 Computational Simulation of a Water-Cooled Heat Pump

15 Computational Model of Heat Transfer on the ISS

15 Optimization of Angular-Momentum Biases of Reaction Wheels

15 Short- and Long-Term Propagation of Spacecraft Orbits

P H Y S I C A L S C I E N C E S

18 Monte Carlo Simulation to Estimate Likelihood of DirectLightning Strikes

18 Adaptive MGS Phase Retrieval

19 Predicting Boundary-Layer Transition on Space-ShuttleRe-Entry

20 Calculations for Calibration of a Mass Spectrometer

20 Simulating the Gradually Deteriorating Performance ofan RTG

20 2D/3D Synthetic Vision Navigation Display

21 Automated Camera Array Fine Calibration

22 Multichannel Networked Phasemeter Readout andAnalysis

22 MISR Instrument Data Visualization

22 Platform for Postprocessing Waveform-Based NDE

23 System for Continuous Delivery of MODIS Imagery toInternet Mapping Applications

24 Automatic Rock Detection and Mapping from HiRISEImagery

24 Parallel Computing for the Computed-TomographyImaging Spectrometer

24 Processing LiDAR Data to Predict Natural Hazards

I N F O R M A T I O N S C I E N C E S

25 Rock Segmentation Through Edge Regrouping

26 DSN Array Simulator

27 Estimating Software-Development Costs With GreaterAccuracy

27 Visual Target Tracking on the Mars Exploration Rovers

28 Parametric-Studies and Data-Plotting Modules for the SOAP

29 Testing of Error-Correcting Sparse Permutation ChannelCodes

30 SPICE Module for the Satellite Orbit Analysis Program(SOAP)

30 Enhanced Reporting of Mars Exploration Rover Telemetry

31 Facilitating Analysis of Multiple Partial Data Streams

32 Service-Oriented Architecture for NVO and TeraGridComputing

32 Mars Reconnaissance Orbiter Wrapper Script

On the Cover:This image, created with COMSOL Multiphysics

software from COMSOL, Inc. (Burlington, MA)

shows a simulation of the thermo-electro-mechanical

effects in an RF coil, visualizing the magnetic flux

lines, thermally induced stress, and deformation.

RF coils come in many shapes and can be found in

such diverse applications as MRI equipment, wireless

devices, and RFID systems. Multiphysics simulation of such components

is critical to understanding failure mechanisms, which typically depend

on a multitude of phenomena such as dielectric breakdown, moisture,

and thermal expansion. (Image courtesy of COMSOL, Inc.)

Cov ToC + – ➭

➮

AIntro

http://www.abpi.net/ntbpdfclicks/l.php?200809NTBHOME

It’s too expensive to develop new technologies in the real world.

At Advanced Computational and Engineering Services, we are a team of

engineers experienced in simulating real world events in a virtual

environment. With our capability in multiphysics analysis, we can identify

the interdependent effects of multiple, simultaneous physical phenomena,

whether they be structural, fluid, acoustic, chemical, electromagnetic or

thermal in nature. This allows you to rapidly screen concepts to increase

the rate – and success – of new product development. It improves

manufacturing processes as well.

Make your technologies a quicker and more profitable reality through our expertise in multiphysics engineering.

Contact us today:(614) 861-7015 www.acescolumbus.com

It’s too expensive to develop new technologies in the real world.

Develop them in ours.

Free Info at http://info.hotims.com/15138-791

Cov ToC + – ➭

➮

AIntro




Design & Analysis Software

The Saab Group has 17 business units,which are split into defense and security,systems and products, and aeronautics.Over the years, the company has takenadvantage of the many paradigm shiftsthat have taken place in engineeringanalysis. One example is implementingcomprehensive engineering methodolo-gies that combine traditional experi-ments and testing with newer tools suchas computer modeling and simulation.In the 1980s, Saab began applying large-scale computer simulations, which wereused to verify the lightning-protectioncomponents in the wings of the Gripenfighter aircraft.

One of the Saab divisions was workingwith the Swedish Defense MaterialAdministration, and engineers in thatdivision were asked to perform a con-ceptual study on what happens to air-plane materials when struck by light-ning. Because weight is a major consid-eration in aeronautic design, these wingsare made of lightweight, yet strong, com-posite materials. These materials aremade up of several layers of different

composites, and in these layers, thematerials often have a different orienta-tion to increase strength. But, becausemodern composites exhibit stronglyanisotropic electrical and thermal con-ductivities, and because they have lowconductivity compared to metals, whenthe high electric currents due to a light-ning strike flow through them, theyexperience a high temperature rise andare vulnerable to heating damage. Theheat flowing through the compositestructure also has an effect on aircraftparts close to the location of the strike.

The anisotropic, layered nature ofthese composites demands a 3D analysis.In addition, the underlying physics arestrongly coupled because the heating,and thus temperature, depends on thecurrent distribution, which in turn isinfluenced by the fact that the compos-ites’ electrical conductivity is tempera-ture-dependent. Any attempt to analyzethe temperature rise becomes a non-triv-ial multiphysics problem.

In initial attempts at modeling thiseffect, engineers tried manipulating in-

house codes and commercial codes toinclude these multiphysics phenomena.However, this proved extremely difficultbecause none of the codes were built forsimultaneously solving the electromag-netic and temperature fields together.

In 2002, Saab learned of COMSOLMultiphysics, a simulation tool whosefundamental structure was built aroundcoupling physics and solving themtogether easily and intuitively. This was,at the time, almost unheard of in codesfor electromagnetic simulations, whichbasically analyzed just the electromag-netic fields; if other physics were to beinvolved in the modeling application,their effect had to be integrated in anempirical or approximate fashion.

The software was built around thephysics coupling approach, making themodeling of lightning strikes on an air-plane easy and affordable. In particu-lar, modeling this effect with COMSOLMultiphysics is possible because of thesoftware’s ability to solve virtually anyset of coupled differential equations.In addition, other physical effects can

Multiphysics Modeling Simplifies Analysis of Electrical andElectromagnetic EffectsEffects of lightning strikes and electromagnetic fields are simulated and modeled easily.Saab Group, Linköping, Sweden, and COMSOL, Inc., Burlington, Massachusetts

Figure 1. Model of Lightning Striking an Airplane Wing. On the left, the slice plot shows current density and the streamlines show current path. On theright, the slice plot shows the temperature, and the boundary plot shows the electric potential from the lightning strike.

Cov ToC + – ➭

➮

AIntro


Imagine being able to design unhindered by the constraints of traditional geometry! SpaceClaim puts

creativity back into design with simple yet powerful “click-and-do” software. It offers a streamlined set of

tools that allows anyone to modify, redesign, or run “what-if” scenarios on 3D models from every major

CAD system. It helps remove design-to-analysis barriers, deliver smarter models for evaluation and

manufacturing, and brings products to market faster at lower costs. SpaceClaim is an ideal workflow

collaboration tool that helps make design and analysis teams super efficient and highly productive.

Design the way you think TM

Seeing is believing. Contact us today for a FREE TRIAL and Whitepaper download.

SpaceClaim’s powerful 3D productivity tool lets you design the way you think.

www.SpaceCla im.com +1 . 9 7 8 . 4 8 2 . 214 8

Introducing freedom of design without constraints.

SPC002_FreedomAdFNL.indd 1 8/11/08 11:25:37 AM

SpaceCla

ful 3D products poweraim’’s

tivity tool lets you design t

the way you think.

Imagine bein

eativity bacr

tools that al

CAD system

manufacturi

collaboration

S

ed by theng able to design unhinder

ack into design with simple yet powe

edesign, or, rllows anyone to modify

m. emove design-to-analysisIt helps r

oducts to market fng, and brings pr

n tool that helps make design and a

Seeing is believing. Contact us tod

ye constraints of traditional geometr

fee. It offul “click-and-do” softwarer

un “what-if” scenarios on 3D modr r

ter models foriers, deliver smars bar

faster at lower costs. SpaceClaim is

analysis teams super efficient and h

day for a FREE TRIAL and Whitepa

y! SpaceClaim puts

eamlined set of rs a str

dels y major om everryfrro

or evaluation and

s an ideal workflow

oductive. highly pr

aper download.

w

SPC002_FrreedomAAdFNL.inddSPC002_FreedomAdFNL.indd

.www SpaceCla im moc. +1

1 1

4 818 . 4 8 2 . 27. 91

Design

n the way you think TM

88/11/0/08 11:25:37 AM8/11/08 11:25:37 AM


Cov ToC + – ➭

➮

AIntro




be added, such as wind cooling andblack-body radiation heating that arisesfrom the hot lightning channel, whichis the 1-cm thick channel of ionizedhot air (10,000-20,000°C) where thelightning discharge flows onto an air-plane wing.

Figure 1 shows the results of one suchsimulation of the heating caused by acurrent pulse from lightning injectedinto the leading edge of a wing that con-sists of two layers of different anisotropicand homogeneous composite materials.The current is injected across a small cir-cular area in the front. The figure showsthe distribution of current density on anumber of vertical and horizontal slices

through the structure at an instant intime just after the lightning has struck.In the left image, the slice plot showscurrent density, while streamlines indi-cate the current’s path. In the figure onthe right, the slice plot describes thetemperature, and the boundary plot inthe middle of the geometry shows theelectric potential.

The figure on the right shows wherethe temperature distribution reachesthe material’s melting temperature of300°C. It is evident that the outer mate-rial layer, which has the lowest electricalconductivity, is severely damaged by thetemperature rise while the inner layer isnot. Furthermore, it is easy to study how

the extent of the damage is influencedby the degree of material anisotropies.

This methodology was validated forsimulating lightning strikes against actualtest results. Radiative heating from thelightning channel also plays an importantrole. The findings from these simulationshad a major impact on construction tech-niques and provided useful design rulesfor the next generation of advancedmaterials for aircraft structures.

Electromagnetic simulations havebeen used to analyze a variety of otheraircraft-related applications such asantenna diagrams, antenna-to-antennacouplings, radar cross-sections, interfer-ence propagation, printed circuit boarddesigns, and test setup optimization.

Saab Group also used the software foran external customer, ABB, in whichcase they modeled the electromagneticeffects on the casing (the electromag-netic shielding) surrounding a voltagesubstation. These electricity distributionsystems are used to transform voltagesbetween different forms and levels andthus provide the link between high-volt-age transmission lines and the domesticelectricity supply.

Substations contain many compo-nents such as switches, transformers,and reactor coils that generate electro-magnetic fields. The strong fields ema-nating from the transformers must fre-quently be shielded so as to protectother equipment and systems in andaround the substation. In this process,however, the enclosure or shield wallsare subject to eddy currents, which canheat the wall material enough to lead tomelting.

The multiphysics coupling betweenheat and electromagnetics was modeled

Figure 3. Model of Three Current-Carrying Coils of different phase (left) and the inductive heating in the electromagnetic shield (right).

Figure 2. COMSOL Multiphysics allows users to Implement an Expression for a Conducting Layer so thesoftware treats a 3D structure as a 2D surface, but nonetheless simulates the 3D behavior. This can beuseful for simulating thin internal borders, such as possible shielding layers modifying the near-fieldfrom a cellular phone.


Cov ToC + – ➭

➮

AIntro


Software Tech Briefs, September 2008 www.techbriefs.com 7

with COMSOL, but in this model, a dif-ferent problem arose. An importantparameter in electromagnetic shieldingis the ratio of the layer thickness, d, tothe penetration (skin) depth, δ. In manysituations, d ≥ δ, particularly at higherfrequencies or for very thick layers.

The finite element method (FEM) iswell suited for modeling arbitrary shapesand coupled phenomena, but it oftenrequires a very fine mesh if it is to resolvethe interior of very thin structures suchas a metal wall, as in the case of theseshields. With standard grid shapes, mod-eling such walls and other thin conduct-ing layers in three dimensions oftenleads to an excessive number of meshelements.

One approach to reduce the numberof elements is to work with scaled orelongated objects, but in many casesthis still leads to a number of elementsthat is difficult or slow to handle. Thesoftware enables implementation of anexpression for the conducting layer,and while it treats the 3D structure as a2D surface, it simulates the layer’s 3Dbehavior. To include the influence of

the layer on the electromagnetic fieldsin the surrounding 3D domain, appro-priate boundary conditions wereapplied across the surface (Figure 2).Thus, while significantly reducing theamount of memory needed and thesolution time by treating the wall layeras a 2D boundary, engineers could stillsimulate the substation enclosure’sinductive wall heating and the shield-ing efficiency.

These methods have been applied tosimulate such cases within microwavephenomena and electromagnetic com-patibility, where an equation at theboundary replaces the need to modelthe thin domain. An added advantage isthat this system of equations can alsosimulate internal borders such asshielding layers modifying the near-field in a cellular phone such asbetween the antenna and other compo-nents (Figure 2).

In this specific application, a modelwas developed of an enclosed substationwith three current-carrying coilsdesigned to reduce reactive power; thatis, to minimize the phase shift between

current and voltage. In this situation, thecurrents induced in the wall are verystrong, leading to high temperatures. Inparticular, current density in the regionsnear openings and slits can become sohigh that the temperature reaches themetal’s melting point.

The model results show that theheating is greatest around the portholeat the front of the electromagneticshield. Figure 3 (left) shows simulationresults for three current-carrying coilsof different phases and reveals the sizeand direction of the magnetic flux.The figure on the right illustratesinductive heating in the electromag-netic shield.

The model uses aluminum as theshielding material, and the results con-firm that heating is greatest around theporthole at the front of the electromag-netic shield. Adjustments in the designare likely necessary in order to reducethe maximum temperature.

This work was done by Dr. Göran Erikssonof the Saab Group, using software from COM-SOL. For more information, visithttp://info.hotims.com/15138-122.

The propagation of shear (S) andcompression (P) waves within the Earthrepresents a critically important phe-nomenon for geologists. For many years,geologists have developed specializedcomputational programs to calculatewave propagation within complex geo-physical regions. These programs havebeen instrumental in determining thelocation and characteristics of naturalphenomena (e.g., earthquakes) andmanmade activity (e.g., nuclear-blasttests).

Since these waves typically travel longdistances prior to detection, these pro-grams typically require large computa-tional models and significant computa-tional resources. Newer applications forthis technology include border securityand exploration for natural resources.The length scales of these applicationsare much smaller than for typical appli-cations. Thus, the P and S waves decayless than in applications with longerlength scales, and therefore cannot beignored. As these industries apply exist-

Geophysical Wave Propagation Calculation Using MultiphysicsGeologists can conduct in-field calculations without the need for large computing resources.Advanced Computational and Engineering Services (ACES), Gahanna, Ohio, and COMSOL, Inc., Burlington, Massachusetts

Seismic Wave propagating through geological domain.

Cov ToC + – ➭

➮

AIntro

http://www.abpi.net/ntbpdfclicks/l.php?200809STBP7



ing technology to meet today’s chal-lenges, they are finding that new methodsof solving these problems have significantadvantages over traditional methods.

New methods are being developed byACES to provide practical solutions togeophysical wave propagation problemsof interest to the Department ofHomeland Security, the Department ofDefense, and the energy industry. Theseindustries, as well as many others, arebeing served by the application of theadvanced computational methods inCOMSOL Multiphysics. These analyses

reduce the need for large computation-al resources and permit geologists toconduct in-field calculations.

The example in the figure shows thevelocity distribution in a half-space sub-jected to a dynamic excitation near theEarth’s surface. The Cartesian plots showthe surface and subsurface response. Forthe simplified case of a homogeneoushalf-space, engineers implementedclosed-form solutions for a wide range ofloadings against which to compare thecomputational results. The forcing func-tions analyzed in this work represent

impact loadings with duration of 10 ms.Thus, element size and time step areimportant model parameters for the accu-rate calculation of the displacement andvelocity associated with the excitation.

During these developments ACES hasshown that COMSOL Multiphysics pro-vides results that accurately representclosed-form and experimental data.

This work was performed by Dr. S.P.Yushanov, Dr. J.S. Crompton, and Dr. K.C.Koppenhoefer of ACES using COMSOL soft-ware. For more information, visithttp://info.hotims.com/15138-123.

Simulating the Manufacturing Process of Ceramic MatrixCompositesMultiphysics simulation enables analysis of a range of physical phenomena occurring in themanufacturing process.Advanced Computational and Engineering Services (ACES), Gahanna, Ohio, and COMSOL, Inc., Burlington,Massachusetts

Increasing the temperature at whichjet aircraft engines operate would signif-icantly improve thrust and fuel efficien-cy with reduced emissions. However, cur-rent engines operate within 50 degreesof the inherent melting point of the con-ventional materials used in engine con-struction. Thus, new materials capableof operating at higher temperatures forprolonged times must be developed andmanufactured.

Ceramics and ceramic matrix compos-ites (CMCs) can operate at temperatures

in excess of 2000°F but are difficult tofabricate into the complex shapesrequired for jet engine use and conse-quently, novel manufacturing processesmust be developed and processing con-ditions optimized for routine produc-tion of complex components.

To support the development of theseinnovative manufacturing processes,ACES has applied COMSOL Multi -physics to the simulation of manufactur-ing processes for the production ofCMCs. The software has been used to

develop specialized multiphysics simula-tion technologies that describe the infil-tration of molten material into a ceram-ic preform. The range of physical phe-nomena considered in the analysis islarge and includes the following criticalmechanisms:• Unsaturated flow of fluid into a ceram-

ic matrix.• Capillary fluid flow.• Chemical reaction between the fluid

and the ceramic matrix.• Volumetric changes associated with

the fluid-solid reaction.• Temperature changes associated with

the fluid-solid reaction.• Residual stress development and its

effect on component shape.COMSOL Multiphysics combines the

native capability to conduct fluid flowand structural mechanics solutions withthe high degree of flexibility necessaryto include the required fluid flow, chem-ical reaction, and thermal calculationsto simulate the critical components ofthe CMC manufacturing process. Partialdifferential equations (PDEs) unique tothis process have been directly pro-grammed into the software to describethe flow of liquid material into a ceram-ic matrix. These equations are directlyincorporated with COMSOL Multi -physics structural and thermal analysesthat are key components of the analysis.The resulting simultaneous solution ofmultiple physical phenomena provides amore accurate analysis of the processDistribution of the Velocity in the fill direction for both layers at t=1s.


Cov ToC + – ➭

➮

AIntro




and allows simulation of the interde-pendent physical phenomena found inthe manufacturing process.

Application of simulation tools of thistype has allowed designers to reducecycle time, increase part yield, and opti-mize the process window for CMC man-

ufacturing. The results of analyses usingCOMSOL Multiphysics have allowedACES to resolve production issues withnew designs prior to mass productionand significantly reduce the time andcost of new product development andmanufacture.

This work was performed by Dr. S.P.Yushanov, Dr. J.S. Crompton, and Dr. K.C.Koppenhoefer of ACES using COMSOL soft-ware. For more information, visithttp://info.hotims.com/15138-124.

3D Program Aids Design of Custom Iron WorkSoftware allows architectural iron works manufacturer to simplify the design process.Keuka Studios, Honeoye Falls, New York, and SpaceClaim, Concord, Massachusetts

Keuka Studios designs and manufac-tures custom architectural iron work andcable railings for commercial and homeuse. Keuka Studios first designs a productin CAD, CNC machines it to shape, andthen takes the assembly to their forge,which adds the design elements. DanWhite, founder and president of KeukaStudios, was frustrated by the limitations ofexisting design tools. Despite over 25 yearsof experience as a mechanical engineerwith a deep working knowledge of 3DCAD, animation, rendering, and finite ele-ment analysis software, he could not be ascreative with product design as he wanted.

Implementing 3D design softwarecalled SpaceClaim, Keuka Studios wasable to use upfront design capabilitiesto become creative with designs. Thefirm combines SpaceClaim’s softwarewith hand-forged, thousand-year-oldprocesses.

The software was used to work directlywith the model using the Pull, Move, Fill,and Combine functions. These tools pro-vide the ability to scale surfaces andsolids to an exact value, combine multi-ple components, and create and editimported sheet metal designs forupstream and downstream applications.This approach lets users edit importeddata directly, without regard to where itoriginated. Designers can work as they’rethinking, making changes on the fly,experimenting with different approach-es and seeing the results immediately.

The Sheet Metal Module provides 3Ddesign and optimization for creatingsheet metal parts and assemblies. Userscan design from scratch or import andedit sheet metal designs. Non-sheetmetal parts also are converted to sheet

metal parts, which can then beunfolded and manufactured. A sin-gle sheet metal part can be split intomultiple parts at critical junctionareas.

SpaceClaim’s modeling tools workin arbitrary cross-section views, draw-ings, and other 3D views of a part orassembly. Users can work in theirfamiliar 2D design views by starting alayout or rotationally symmetricdesign in 2D, and then see the 3Dviews.

The software was able to updatethe model immediately, and behavethe way the user intended it to. Moreimportantly, the software is not basedon history. Users can slice and dicethe model, enabling them to focuson creativity in an unconstrainedenvironment without thinking abouthow their design will function.According to White, the softwareprovides an environment that makesthe 2D experience as easy as 3D.

With traditional design programs,designers first have to plan every-thing out, and then do conceptualwork sketching on paper and think-ing outside the CAD system. If they

bypass this “out of the CAD” system plan-ning, the system often penalizes them.The software enabled White to chop andmanipulate the designs as they evolved.

For more information, visit http://info.hotims.com/15138-125.

Figure 1. Using SpaceClaim, the width of the legs ofa bench is determined. The software mirrors whatchanges are made to one side of the bench, andautomatically changes the other side.

Figure 2. The bench in its final form. It was designed inSpaceClaim and fabricated by Keuka Studios. A bronzepatina was applied for the finish.

Displaying CFD Solution Parameters on Arbitrary Cut PlanesLangley Research Center, Hampton, Virginia

USMC6 is a Fortran 90 computer pro-gram for post-processing in support ofvisualization of flows simulated by com-putational fluid dynamics (CFD). The

name “USMC6” is partly an abbreviationof “TetrUSS — USM3D Solution Cutter,”reflecting its origin as a post-processorfor use with USM3D — a CFD program

that is a component of the TetrahedralUnstructured Software System and thatsolves the Navier-Stokes equations ontetrahedral unstructured grids. “Cutter”

Cov ToC + – ➭

➮

AIntro





here refers to a capability to acquire andprocess solution data on (1) arbitraryplanes that cut through grid volumes, or(2) user-selected spheroidal, conical,cylindrical, and/or prismatic domainscut from within grids. Cutting saves timeby enabling concentration of post-pro-cessing and visualization efforts onsmaller solution domains of interest.

The user can select from among morethan 40 flow functions. The cut planescan be trimmed to circular or rectangu-lar shape. The user specifies cuts andfunctions in a free-format input fileusing simple and easy-to-remember key-words. The USMC6 command line issimple enough that the slicing processcan readily be embedded in a shell script

for assembly-line post-processing. Theoutput of USMC6 is a data file ready forplotting.

This program was written by S. Paul Pao ofLangley Research Center. For more informa-tion, download the Technical Support Package(free white paper) at www.techbriefs.com/tspunder the Software category. LAR-17527-1

This software solves the Navier-Stokes equations for the incompress-ible driven cavity flow problem. Thecode uses second-order finite differ-encing on a staggered grid using theChorin projection method. The result-ing intermediate Poisson equation is

efficiently solved using the fast Fouriertransform.

Time stepping is done using fourth-orderRunge-Kutta for stability at high Reynoldsnumbers. Features include check-pointing,periodic field snapshots, ongoing reportingof kinetic energy and changes between time

steps, time histories at selected points, andoptional streakline generation.

This program was written by Virginia Kalbof Goddard Space Flight Center. For furtherinformation, contact the Goddard InnovativePartnerships Office at (301) 286-5810. GSC-15107-1

Flow Solver for Incompressible 2-D Drive CavityGoddard Space Flight Center, Greenbelt, Maryland

Flow Solver for Incompressible Rectangular DomainsGoddard Space Flight Center, Greenbelt, Maryland

This is an extension of the FlowSolver for Incompressible 2-D DriveCavity software described in the pre-ceding article. It solves the Navier-Stokes equations for incompressibleflow using finite differencing on a uni-form, staggered grid. There is a run-time choice of either central differenc-ing or modified upwinding for the con-vective term. The domain must be rec-tangular, but may have a rectangularwalled region within it. Currently, theposition of the interior region and

exterior boundary conditions arechanged by modifying parameters inthe code and recompiling. These fea-tures make it possible to solve a varietyof classical fluid flow problems such asan L-shaped cavity, channel flow, orwake flow past a square cylinder. Thecode uses fourth-order Runge-Kuttatime-stepping and overall second-orderspatial accuracy.

This software permits the walledregion to be positioned such that flowpast a square cylinder, an L-shaped cavi-

ty, and the flow over a back-facing stepcan all be solved by reconfiguration.Also, this extension has an automaticdetection of periodicity, as well as use ofspecialized data structure for ease ofconfiguring domain decomposition andcomputing convergence in overlapregions.

This program was written by Virginia L.Kalb of Goddard Space Flight Center. For fur-ther information, contact the GoddardInnovative Partnerships Office at (301) 286-5810. GSC-15111-1

Simulating Avionics Upgrades to the Space ShuttlesLyndon B. Johnson Space Center, Houston, Texas

Cockpit Avionics Prototyping En vi -ronment (CAPE) is a computer programthat simulates the functions of proposedupgraded avionics for a space shuttle. InCAPE, pre-existing space-shuttle-sim -ulation programs are merged with a

commercial-off-the-shelf (COTS) dis-play-development program, yielding apackage of software that enables high-fidelity simulation while making it possi-ble to rapidly change avionic displaysand the underlying model algorithms.

The pre-existing simulation programsare Shuttle Engineering Simulation,Shuttle Engineering Simulation II,Inter active Control and DockingSimulation, and Shuttle Mission Sim -ulator playback.

Electronics/Computers


Cov ToC + – ➭

➮

AIntro

http://www.abpi.net/ntbpdfclicks/l.php?200809NTBPTSP



Simulating the Phoenix Landing Radar SystemNASA’s Jet Propulsion Laboratory, Pasadena, California

Injecting Artificial Memory Errors Into a Running ComputerProgramNASA’s Jet Propulsion Laboratory, Pasadena, California

The COTS program — Virtual Ap pli -cation Prototyping System (VAPS) — notonly enables the development of displaysbut also makes it possible to move dataabout, capture and process events, andconnect to a simulation. VAPS alsoenables the user to write code in the C orC++ programming language and compile

that code into the end-product simulationsoftware. As many as ten different avionic-upgrade ideas can be incorporated in asingle compilation and, thus, tested in asingle simulation run. CAPE can be run inconjunction with any or all of four simula-tions, each representing a different phaseof a space-shuttle flight.

This program was written by Daniel Degerand Kenneth Hill of Johnson Space Centerand Karsten E. Braaten of United SpaceAlliance. For more information, downloadthe Technical Support Package (free whitepaper) at www.techbriefs.com/tsp under theSoftware category. MSC-23453-1/15-1

A computer program called “phxlr-sim” simulates the behavior of the radarsystem used as an altimeter and vel -ocimeter during the entry, descent, andlanding phases of the Phoenix landerspacecraft. The simulation includesmodeling of internal functions of theradar system, the spacecraft trajectory,and the terrain. The computationalmodels incorporate representations ofnonideal hardware effects in the radarsystem and effects of radar speckle(coherent scatter of radar signals fromterrain).

This program was written by Curtis W.Chen of Caltech for NASA’s Jet PropulsionLaboratory. For more information, down-load the Technical Support Package (freewhite paper) at www.techbriefs.com/tspunder the Software category.

This software is available for commer-cial licensing. Please contact KarinaEdmonds of the California Institute ofTechnology at (626) 395-2322. Refer toNPO-44431. Phoenix Mission Lander on Mars, artist’s concept.

Single-event upsets (SEUs) or “bit-flips” are computer memory errorscaused by radiation. BITFLIPS (BasicInstrumentation Tool for FaultLocalized Injection of ProbabilisticSEUs) is a computer program that delib-erately injects SEUs into another com-puter program, while the latter is run-ning, for the purpose of evaluating thefault tolerance of that program. BIT-FLIPS was written as a plug-in extensionof the open-source Valgrind debuggingand profiling software. BITFLIPS caninject SEUs into any program that can

be run on the Linux operating system,without needing to modify the pro-gram’s source code. Further, if access tothe original program source code isavailable, BITFLIPS offers fine-grainedcontrol over exactly when and whichareas of memory (as specified via pro-gram variables) will be subjected toSEUs.

The rate of injection of SEUs is con-trolled by specifying either a fault proba-bility or a fault rate based on memorysize and radiation exposure time, in unitsof SEUs per byte per second. BITFLIPS

can also log each SEU that it injects and,if program source code is available,report the magnitude of effect of theSEU on a floating-point value or otherprogram variable.

This program was written Benjamin J.Bornstein, Robert A. Granat, and Kiri L.Wagstaff of Caltech for NASA’s Jet PropulsionLaboratory.

This software is available for commerciallicensing. Please contact Karina Edmonds ofthe California Institute of Technology at(626) 395-2322. Refer to NPO-45368.

Cov ToC + – ➭

➮

AIntro




Firmware for implementing a digitalphasemeter within a field-programmablegate array (FPGA) has been devised. Inthe original application of this firmware,the phase that one seeks to measure isthe difference between the phases of twonominally-equal-frequency heterodyne

signals generated by two interferome-ters. In that application, zero-crossingdetectors convert the heterodyne signalsto trains of rectangular pulses (see fig-ure), the two pulse trains are fed to afringe counter (the major part of thephasemeter) controlled by a clock signal

having a frequency greater than the het-erodyne frequency, and the fringe count-er computes a time-averaged estimate ofthe difference between the phases of thetwo pulse trains.

The firmware also does the following:• Causes the FPGA to compute the fre-

quencies of the input signals;• Causes the FPGA to implement an

Ethernet (or equivalent) transmitterfor readout of phase and frequencyvalues; and

• Provides data for use in diagnosis ofcommunication failures.The readout rate can be set, by pro-

gramming, to a value between 250 Hzand 1 kHz. Network addresses can beprogrammed by the user.

This program was written by Shanti R.Rao of Caltech for NASA’s Jet PropulsionLaboratory. For more information, down-load the Technical Support Package (freewhite paper) at www.techbriefs.com/tspunder the Software category.

The software used in this innovation isavailable for commercial licensing. Pleasecontact Karina Edmonds of the CaliforniaInstitute of Technology at (626) 395-2322.Refer to NPO-45575.


The Firmware Code converts two inputs (reference and measure) into a time-averaged estimate of thephase difference between the two signals.

Implementing a Digital Phasemeter in an FPGANASA’s Jet Propulsion Laboratory, Pasadena, California

Fault-Tolerant, Multiple-Zone Temperature ControlNASA’s Jet Propulsion Laboratory, Pasadena, California

A computer program hasbeen written as an essentialpart of an electronic tempera-ture control system for aspaceborne instrument thatcontains several zones. Thesystem was developed becausethe temperature and the rateof change of temperature ineach zone are required to bemaintained to within limitsthat amount to degrees of pre-cision thought to be unattain-able by use of simple bimetal-lic thermostats.

The software collects tem-perature readings from sixplatinum resistance ther-mometers, calculates temper-ature errors from the read-ings, and implements a pro-portional + integral + deriva-

tive (PID) control algorithmthat adjusts heater power levels.The software accepts, via a serialport, commands to change itsoperational parameters. Thesoftware attempts to detect andmitigate a host of potentialfaults. It is robust to many kindsof faults in that it can maintainPID control in the presence ofthose faults (see figure).

This program was written byJames Granger, Brian Franklin,Martin Michalik, Phillip Yates, ErikPeterson, and James Borders ofCaltech for NASA’s Jet PropulsionLaboratory.

This software is available for com-mercial licensing. Please contactKarina Edmonds of the CaliforniaInstitute of Technology at (626) 395-2322. Refer to NPO-45230.Logical Architecture of the active thermal control software.

Startup Tasks

Read Temperature

Filter Readings

PID Calculation

Generate Telemetry

Command New Power

Fault Correction

Command Processing

Watchdog Handling

f2

f1

(f1-f2)

Fringe

counter

Reference

Measure

Clock

Zero-crossing

detectors

Phasemeter

To the customer’s data acquisition system

Light

sources

Reference

Measure

Electronics/Computers

Cov ToC + – ➭

➮

AIntro




Post-Flight Estimation of Motion of Space Structures: Part 1NASA’s Jet Propulsion Laboratory, Pasadena, California

A computer program estimates the rela-tive positions and orientations of two spacestructures from data on the angular posi-tions and distances of fiducial objects onone structure as measured by a target-tracking electronic camera and laser rangefinders on another structure. The pro-gram is written specifically for determin-ing the relative alignments of two anten-nas, connected by a long truss, deployed inouter space from a space shuttle.

The program is based partly on trans-formations among the various coordi-

nate systems involved in the measure-ments and on a nonlinear mathematicalmodel of vibrations of the truss. Theprogram implements a Kalman filterthat blends the measurement data withdata from the model. Using time seriesof measurement data from the trackingcamera and range finders, the programgenerates time series of data on the rela-tive position and orientation of theantennas. A similar program describedin a prior NASA Tech Briefs article wasused onboard for monitoring the struc-

tures during flight. The present pro-gram is more precise and designed foruse on Earth in post-flight processing ofthe measurement data to enable correc-tion, for antenna motions, of scientificdata acquired by use of the antennas.

This program was written by PaulBrugarolas and William Breckenridge ofCaltech for NASA’s Jet Propulsion Laboratory.


Post-Flight Estimation of Motion of Space Structures: Part 2NASA’s Jet Propulsion Laboratory, Pasadena, California

A computer program related to theone described in the immediately pre-ceding article estimates the relative posi-tion of two space structures that arehinged to each other. The input to theprogram consists of time-series data ondistances, measured by two range find-ers at different positions on one struc-ture, to a corner-cube retroreflector onthe other structure. Given a Cartesian(x,y,z) coordinate system and the knownx coordinate of the retroreflector rela-

tive to the y,z plane that contains therange finders, the program estimates they and z coordinates of the retroreflector.

The estimation process involves solv-ing for the y,z coordinates of the inter-section between (1) the y,z plane thatcontains the retroreflector and (2)spheres, centered on the range finders,having radii equal to the measured dis-tances. In general, there are two suchsolutions and the program chooses theone consistent with the design of the

structures. The program implements aKalman filter. The output of the pro-gram is a time series of estimates of therelative position of the structures.

This program was written by PaulBrugarolas and William Breckenridge ofCaltech for NASA’s Jet Propulsion Laboratory.

This software is available for commerciallicensing. Please contact Karina Edmonds ofthe California Institute of Technologyat (626) 395-2322. Refer to NPO-45074.

The Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) is a computer program for sim-ulating transient operation of a com-mercial turbofan engine that can gener-ate as much as 90,000 lb (≈0.4 MN) ofthrust. It includes a power-managementsystem that enables simulation of open-or closed-loop engine operation over awide range of thrust levels throughoutthe full range of flight conditions.

C-MAPSS runs in the Simulink (TheMathworks, Inc.) block-diagram lan-

guage, providing a graphical simulationenvironment in which advanced controland diagnostics algorithms can be imple-mented and tested. The software has agraphical user interface (GUI) thatmakes engine “health” data and controland engine parameters easily accessible.It can run user-specified transient simu-lations and generate state-space linearmodels of a nonlinear engine model atan operating point.

C-MAPSS produces GUI screens thatenable point-and-click operation and

include editable fields for user-specifiedinput. The software includes an atmos-pheric model for simulating operation ataltitudes from sea level to 40,000 ft (≈12km), Mach numbers from 0 to 0.90, andsea-level ambient temperatures from –60to +103 °F (≈–51 to +39 °C). CMAPSS hasa comprehensive control system consist-ing of a gain-scheduled fan-speed con-troller and several limit regulators, inte-grated in a manner similar to that usedin real engine controllers to avoid inte-grator windup. The simulation code

Simulating Operation of a Large Turbofan EngineJohn H. Glenn Research Center, Cleveland, Ohio

Mechanics/Machinery

Cov ToC + – ➭

➮

AIntro



itself operates several times faster thanreal time, giving it the potential to bedeployed (all or in part) as machinecode for hardware-in-the-loop applica-tions such as flight simulators and real-time controller/diagnostic system valida-tion.

Overall, C-MAPSS provides the userwith a set of tools for performing open-and closed-loop transient simulations

and comparison of linear and non-linearmodels throughout its operating enve-lope, in an easy-to-use graphical environ-ment.

This program was written by Jonathan S.Litt of Glenn Research Center; Dean K.Frederick of Saratoga Control Systems, Inc.;and Jonathan A. DeCastro of ASRC AerospaceCorp. For more information, download theTechnical Support Package (free white

paper) at www.techbriefs.com/tsp under theSoftware category.

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, InnovativePartnerships Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road,Cleveland, Ohio 44135. Refer to LEW-18315-1.

Automated Assistance for DesigningActive Magnetic BearingsStennis Space Center, Mississippi

MagBear12 is a computer code (seefigure) that assists in the design of radi-al, heteropolar active magnetic bearings(AMBs). MagBear12 was developed tohelp in designing the system describedin “Advanced Active-Magnetic-BearingThrust-Measurement System” (SSC-00177-1), which appears in NASA TechBriefs, Vol. 32, No. 9 (September 2008), p.61.. (See the Mechanics/Machinery sec-tion in the accompanying issue of NASATech Briefs). Beyond this initial applica-tion, MagBear12 is expected to be use-ful for designing AMBs for a variety ofrotating machinery. This programincorporates design rules and govern-ing equations that are also implement-ed in other, proprietary design softwareused by AMB manufacturers. In addi-tion, this program incorporates anadvanced unpublished fringing-mag-netic-field model that increases accura-cy beyond that offered by the otherAMB-design software.

MagBear12 accepts input from the userin the form of parameters that specify theenvelope, performance, and acceptableranges of geometric features other thanthe envelope. The program then calcu-lates optimized designs within thoseranges. A series of designs are presentedto the designer for review. The designercan accept one of the designs or can mod-ify the input parameters to refine thedesigns. The program can also be used toanalyze pre-existing AMB designs.

This work was done by Joseph Imlach ofInnovative Concepts In Engineering LLC forStennis Space Center.

Inquiries concerning rights for the commercialuse of this invention should be addressed to:

Innovative Concepts In Engineering LLC2142 Tributary CircleAnchorage, AK 99516(907) 337-8954Refer to SSC-00176-1, volume and number

of this NASA Tech Briefs issue, and thepage number. MagBear12 Flowchart

Computational Simulation of a Water-Cooled Heat PumpLyndon B. Johnson Space Center, Houston, Texas

A Fortran-language computer pro-gram for simulating the operation of awater-cooled vapor-compression heatpump in any orientation with respect togravity has been developed by modify-ing a prior general-purpose heat-pumpdesign code used at Oak RidgeNational Laboratory (ORNL). Al -though it is specific to the design of ahigh-temperature-lift heat pump forthe International Space Station, thisprogram could serve as a basis fordevelopment of general-purpose com-putational software for designing and

analyzing liquid-cooled heat-pumps.The ORNL program contained modelsof refrigerant-fluid-to-air heat exchang-ers; the main modification consisted inreplacing those models with models ofplate-type heat exchangers utilizingwater as both the cooling and the heat-ing source liquid.

The present program incorporates aFortran implementation of theAmerican Society of Mechanical En -gineers water-properties tables. Semi-empirical models of the heat transfercoefficients for these heat exchangers

were developed from vendor and labo-ratory test data, inasmuch as applicablepublished correlations were not avail-able. The program produces estimatesof evaporator and condenser capaci-ties, coefficients of performance, andoperating temperatures over a range ofcompressor speeds.

This work was done by Duane Bozarth ofH and R Technical Associates for JohnsonSpace Center. For more information, down-load the Technical Support Package (freewhite paper) at www.techbriefs.com/tspunder the Software category. MSC-23375-1

N

Y

Y

N

Get Data From Input Forms

Start

Calculate RequiredWidth of Pole Piece

Determine OptimumCoil Dimensions

Calculate PolePiece Height

Calculate Output Parameters

Display Output

NP = NPmax ?

Modify Input?

Finish

Mechanics/Machinery

Cov ToC + – ➭

➮

AIntro





Computational Model of Heat Transfer on the ISSLyndon B. Johnson Space Center, Houston, Texas

SCRAM Lite (“SCRAM” signifies“Station Compact Radiator AnalysisModel”) is a computer program for ana-lyzing convective and radiative heat-transfer and heat-rejection performanceof coolant loops and radiators, respec-tively, in the active thermal-control sys-tems of the International Space Station(ISS). SCRAM Lite is a derivative ofprior versions of SCRAM but is morerobust.

SCRAM Lite computes thermal oper-ating characteristics of active heat-trans-port and heat-rejection subsystems for

the major ISS configurations from Flight5A through completion of assembly. Theprogram performs integrated analysis ofboth internal and external coolant loopsof the various ISS modules and of anexternal active thermal control system,which includes radiators and the coolantloops that transfer heat to the radiators.The SCRAM Lite run time is of the orderof one minute per day of mission time.The overall objective of the SCRAM Litesimulation is to process input profiles ofequipment-rack, crew-metabolic, andother heat loads to determine flow rates,

coolant supply temperatures, and avail-able radiator heat-rejection capabilities.Analyses are performed for timelines ofactivities, orbital parameters, and atti-tudes for mission times ranging from afew hours to several months.

This program was written by John G.Torian and Michael L. Rischar of UnitedSpace Alliance for Johnson Space Center. Formore information, download the TechnicalSupport Package (free white paper) atwww.techbriefs.com/tsp under the Softwarecategory. MSC-23622-1

RBOT [RWA Bias Optimization Tool(wherein “RWA” signifies “ReactionWheel Assembly”)] is a computer pro-gram designed for computing angular-momentum biases for reaction wheelsused for providing spacecraft pointing invarious directions as required for scientif-ic observations. RBOT is currentlydeployed to support the Cassini missionto prevent operation of reaction wheels atunsafely high speeds while minimizingtime in undesirable low-speed range,where elasto-hydrodynamic lubricationfilms in bearings become ineffective, lead-ing to premature bearing failure. The

problem is formulated as a constrainedoptimization problem in which maximumwheel speed limit is a hard constraint anda cost functional that increases as speeddecreases below a low-speed threshold.

The optimization problem is solvedusing a parametric search routineknown as the Nelder-Mead simplex algo-rithm. To increase computational effi-ciency for extended operation involvinglarge quantity of data, the algorithm isdesigned to (1) use large time incre-ments during intervals when spacecraftattitudes or rates of rotation are nearlystationary, (2) use sinusoidal-approxima-

tion sampling to model repeated longperiods of Earth-point rolling maneu-vers to reduce computational loads, and(3) utilize an efficient equation toobtain wheel-rate profiles as functions ofinitial wheel biases based on conserva-tion of angular momentum (in an iner-tial frame) using pre-computed terms.

This work was done by Clifford Lee andAllan Lee of Caltech for NASA’s Jet PropulsionLaboratory. For more information, downloadthe Technical Support Package (free whitepaper) at www.techbriefs.com/tsp under theSoftware category.NPO-42011

Optimization of Angular-Momentum Biases of Reaction WheelsNASA’s Jet Propulsion Laboratory, Pasadena, California

Short- and Long-Term Propagation of Spacecraft OrbitsNASA’s Jet Propulsion Laboratory, Pasadena, California

The Planetary Observer PlanningSoftware (POPS) comprises four com-puter programs for use in designingorbits of spacecraft about planets. Theseprograms are the Planetary ObserverHigh Precision Orbit Propagator(POHOP), the Planetary ObserverLong-Term Orbit Predictor (POLOP),the Planetary Observer Post Processor(POPP), and the Planetary ObserverPlotting (POPLOT) program.

POHOP and POLOP integrate theequations of motion to propagate aninitial set of classical orbit elements toa future epoch. POHOP models short-term (one revolution) orbital motion;

POLOP averages out the short-termbehavior but requires far less process-ing time than do older programs thatperform long-term orbit propaga-tions.

POPP postprocesses the spacecraftephemeris created by POHOP orPOLOP (or optionally can use a less-accurate internal ephemeris) tosearch for trajectory-related geometricevents including, for example, risingor setting of a spacecraft as observedfrom a ground site. For each suchevent, POPP puts out such user-speci-fied data as the time, elevation, andazimuth.

POPLOT is a graphics program thatplots data generated by POPP. POPLOTcan plot orbit ground tracks on a worldmap and can produce a variety of sum-maries and generic ordinate-vs.-abscissaplots of any POPP data.

This program was written by John C.Smith, Jr., Theodore Sweetser, Min-KunChung, Chen-Wan L. Yen, Ralph B. Roncoli,and Johnny H. Kwok of Caltech, and Mark A.Vincent of Raytheon for NASA’s JetPropulsion Laboratory.


Cov ToC + – ➭

➮

AIntro




www.comsol.comCOMSOL MULTIPHYSICS IS A REGISTERED TRADEMARK OF COMSOL AB. SOLIDWORKS IS A REGISTERED TRADEMARK OF SOLIDWORKS CORPORATION. INVENTOR IS A REGISTERED TRADEMARK OF AUTODESK, INC. PARASOLID IS A REGISTERED TRADEMARK OF SIEMENS PRODUCT

LIFECYCLE MANAGEMENT SOFTWARE INC. ACIS AND SAT ARE REGISTERED TRADEMARKS OF SPATIAL CORP. OTHER PRODUCT OR BRAND NAMES ARE TRADEMARKS OR REGISTERED TRADEMARKS OF THEIR RESPECTIVE HOLDERS. © 2008 COMSOL, INC. ALL RIGHTS RESERVED.

MESH

Automatic mesh generation of triangular,

quadrilateral, and tetrahedral elements

Independent mesh size control for edges,

surfaces, and volumes

Swept and mapped meshing

Interactive meshing for separate

treatment of geometry parts

Parallelized and adaptive meshing

PHYSICS

Predefined interfaces for a wide

range of physics

Rapid multiphysics model set-up

Easy to add new physics when

needed

Material library with temperature

dependent properties

CAD

Built-in CAD tools

Bi-directional interface to SolidWorks®

and Inventor®

CAD import from CATIA®, Pro/E ®,

NX TM, SolidEdge®, and more

Nastran® mesh import

Geometry repair and defeaturing

ECAD import

Streamlining the Mult From COMSOL Multiphysics® modeling tools’ seamless CAD

facilities to its powerful postprocessing, every step reaches for

perfection. Unlimited physics modeling capabilities and high-

performance numerical solvers bring best-in-class accuracy

and speed to your simulations.

CADDAC

Built-in

ecBi-dir

C

orksctional interface to SolidWWo ®

PHYSICSSCISYHP

edefirP

ecBi dir

evnIand

DAC im

XN S,,MT

astranN

eomeG

DACE i

SS

or a widened interfaces f

orksctional interface to SolidWWo

entor®

AITAATCom t frmpor Eo/rP,® ®,

dgeESolid ® eand mor,,

n® t mesh impor

eaturingepair and defy retr

tmpor

range o

pid maR

asy toE

needed

Materia

depend

MM

ysicsof ph

ysics model set-upultiphm

ysics whenw phd ne ad

ey with temperatural librar

tiesoperdent pr

MESHHSEM

utomatic mesh generation of triaA

and tetrahedral elemequadrilateral,,

or ol ffondependent mesh size contrI

olumesand vsurfaces,,

pped meshingept and maSw

or separatee meshing ffonteractivI

tsy pareatment of geometrtr

,ngularr,

ents

edges,

.cwwww.

NEMEAGNAMELCYCEFIL

SCISYHPITLUMLOSMOC

comsol.com

FOSKRAMEDARTDERETSIGERERATASDNASICA.CNIERAWWATFOSTN

ERASISKROWDILOS.BALOSMOCFOKRAMEDARTDERETSIGERASI

AMEDARTERASEMANDNARBROTCUDORPREHTOPP.ROCLAITAATPS

OTNEVNI.NOITAROPROCSKROWDILOSFOKRAMEDARTDERETSIGE

y pg

e meshingptivarallelized and adaP

.SREDLOHEVITCEPSERRIEHTFOSKRAMEDARTDERETSIGERROSKR

ASIDILOSARAPPA.CNI,KSEDOTUAFOKRAMEDARTDERETSIGERASIR

.DEVRESERSTHGIRLLA.CNI,LOSMOC8002©

T CUDORPSNEMEISFOKRAMEDARTDERETSIGERA

Cov ToC + – ➭

➮

AIntro


Free Info at http://info.hotims.com/1 38-793

SOLVE AND POSTPROCESSING

High-performance direct solvers including

multicore parallel solvers

Iterative solvers including geometric multigrid

pre-conditioners

Fully coupled or sequential solution coupling

using the GUI-based solver script

Parametric and segregated solvers

Interactive visualization of any variable or

function in space and time

Virtually unlimited visualization capabilities using

COMSOL Script® programming

Evaluations of arbitrary functions including line,

surface, and volume integrals

% Multiphysicsxfem=multiphysics(xfem);

% Extended meshxfem.mesh=xmesh(xfem);

% Solve problemxfem.sol=femstatic(fem, ... ’solcomp’,{’V’,’tAxAyAz20’,’tAxAyAz21’,’psi’,’tAxAyAz10’}, ... ’outcomp’,{’V’,’tAxAyAz20’,’tAxAyAz21’,’psi’,’tAxAyAz10’}, ... ’nonlin’,’off’, ... ’linsolver’,’gmg’, ... ’rhob’,5, ... ’mgcycle’,’f’, ... ’presmooth’,’vanka’, ...

C>> clear all;

C>> geom=xfem.geom;C>> geomplot(geom);C>> mesh=xfem.mesh;C>> sol=xfem.sol;

geom [1x1 solid3]

mesh [1x1 femmesh]

sol [1x1 femsol]

xfem [1x1 struct]

clear all;xfem=flload(’power_inductor’);edit power_circuitgeom=xfem.geom;geomplot(geom);mesh=xfem.mesh;postplot(fem, ’slicedata’,{’normB_emqav’},’slicexspacing’,1,’tridata’,{’V’},’arrowdata’,{’Bx_emqav’,’By_emqav’,’Bz_

tiphysics Workflow

SCRIPT

Script for extensive parametric studies

Built-in functions for model validation and

calibration

Foreign file interface to C, Fortran, and Java

Optional add-ons for reaction engineering,

optimization, and signal processing

GUI-builder for customized interfaces

SCRIP

PT

eorF

Opti

optim

IUG -

and Jatran,, orF,eign file interface to C

eaction engineerinor rd-ons fional ad

ocessing and signal prmization,,

or customized interfaces-builder f

vaaav

ng,

SO VVE AND POSTPROORPTSOPDNAEVLLOS

ectormance dirigh-perfH

CESSINGGNISSEC

ers includingsolv

% Multixfem=mu

% Extenxfem.me

% Solvexfem.so

C>> cl

iphysicsultiphysics(xfem);

nded meshesh=xmesh(xfem);

e problemol=femstatic(fem, ...

’solcomp’,{’V’,’tAxAyAz20’,’tAxAyAz2’outcomp’,{’V’,’tAxAyAz20’,’tAxAyAz2’nonlin’,’off’, ...’linsolver’,’gmg’, ...’rhob’,5, ...’mgcycle’,’f’, ...’presmooth’,’vanka’, ...

ear all;

21’,’psi21’,’psi

geom [1x1 solid3]

ect ormance dirigh-perfH

erse parallel solvulticorm

ers including e solvterativI

e-conditionerspr

y coupled or sequentiaullF

using the IUG e-based solv

egatedarametric and segrP

e visualization ofnteractivI

function in space and time

y unlimited visualiztuallirV

C LOSMO Script® ogram pr

fl f bE

ers including solv

ultigrid geometric m

al solution coupling

er script

ersd solv

y variable orf an

e

pabilities using zation ca

mming

l d l

C>> geC>> geC>> meC>> so

clear all;xfem=flloedit powgeom=xfgeomplomesh=xfpostplot(

om=xfem.geom;omplot(geom);sh=xfem.mesh;l=xfem.sol;

oad(’power_inductor’);er_circuitem.geom;t(geom);em.mesh;fem, ’slicedata’,{’normB_emqav’},’slicexspacing’,1,’tridata’,{’V’}

mesh [1x1 femmesh]

sol [1x1 femsol]

xfem [1x1 struct]

},’arrowdata’,{’Bx_emqav’,’By_emqav’,’Bz_

y fuvaluations of arbitrarE

olume integrand v,,surface

,unctions including line

rals

51

Cov ToC + – ➭

➮

AIntro



Physical Sciences

Monte Carlo Simulation To Estimate Likelihood of DirectLightning StrikesJohn F. Kennedy Space Center, Florida

A software tool has been designed toquantify the lightning exposure at launchsites of the stack at the pads under differ-ent configurations. In order to predictlightning strikes to generic structures,this model uses leaders whose origins (inthe x–y plane) are obtained from a 2Drandom, normal distribution. The strik-ing distance is a function of the strokepeak current, which is obtained from arandom state machine that extracts thestroke peak current from a lognormal dis-tribution. The height in which the lead-ers are originated is fixed and chosen tobe several “strike distances” greater thanthe tallest object under study.

The Monte Carlo simulation tool usesseveral random state machines to gener-ate x–y origin of leaders and peak stroke

currents. The structures under study areentered in text files whose names areused as descriptors for report purposes.So, “External_tank.txt” could be the textfile that contains all the vertices of theexternal tank. The lines of the text filescontain three points (x, y, and z) thatdefine “points” of lines or “vertices” ofpolygons. A line composed of threezeros (0 0 0) is used to indicate the endof a line or polygon.

Imaginary spheres (whose diametersare the striking distances) are drawn asthe leader descends vertically toground, and the first object intersectedis considered to be struck. Therefore,the last step of the leader can be in anydirection. The leaders can move in thez direction only, or in a random xyz

direction (software selectable). Theleader steps can be either fixed or vari-able. The length of the study is also soft-ware selectable, so the user can per-form a study of “n” number of years. Asummary report generated by the soft-ware indicates the frequency at whichobjects under study will be struck bylightning.

This work was done by Carlos Mata andPedro Medelius of ASRC AerospaceCorporation for Kennedy Space Center. Formore information, contact Pedro Medelius [email protected], (321) 867-6335, Mail Code: ASRC-19, Kennedy SpaceCenter, FL 32899. KSC-12882

Adaptive MGS Phase Retrieval NASA’s Jet Propulsion Laboratory, Pasadena, California

Adaptive MGS Phase Retrieval softwareuses the Modified Gerchberg-Saxton(MGS) algorithm, an image-based sens-ing method that can turn any focal planescience instrument into a wavefront sen-sor, avoiding the need to use externalmetrology equipment. Knowledge of thewavefront enables intelligent control ofactive optical systems.

The software calculates the opticalpath difference (wavefront) errors inthe exit pupil of an optical system usingonly intensity images of a point of light.The light input may be a star, laser, orany point source measured at symmet-ric positions about focus and at thepupil. As such, the software is a keyenabling technology for space tele-scopes. With only a basic understandingof the optical system parameters (e.g.imaging wavelength, f/number, meas-urement positions, etc.), the softwareevolves an internal model of the opticalsystem to best match the data ensem-ble. Once optimized, the software pro-

ceeds to accurately estimate the wave-front of light as it travels through theoptical system.

The MGS software is highly adaptableto a large range of optical systems andincludes many innovative features. Thisversion does not require an extensiveand complete understanding of the sys-tem under test. Instead, using AutomaticModel Adaptation, only the most basicsystem characteristics must be known.The algorithm adapts these parametersto best fit the data ensemble. These stepsare crucial in achieving extremely highaccuracy in the wavefront solution at thesystem exit-pupil. In addition, a conver-gence-monitoring feature allows thealgorithm to stop when the wavefrontsolution has been reached to within aspecified error tolerance level.

The software also facilitates the appli-cation of prior system knowledge to bet-ter deal with high-dynamic range wave-front errors. This is especially importantwhere the error magnitude is much

greater than the imaging wavelength (asignificant problem in wavefront sens-ing). The software can use wavefrontmodels based on previous runs or opticalmeasurements, or predictions fromexternal models, to initiate a prior phaseestimate, through its Prior Phase BuilderGraphical User Interface. The priorphase is treated by the software as aNumerical Nulling Reference, which isevolved in an outer-outer loop duringcomputation, until it contains the fullsolution. The innermost iteration thenhas the simpler job of estimating the low-dynamic range residual difference of thetrue wavefront error from the NullingReference model. This allows the innerloop to operate around null, where it ismost accurate and robust.

In addition to the wavefront solution,the software can provide an improvedset of system parameters. For example,the result can report the true position ofbest focus and true f/number in theoptical system.

Cov ToC + – ➭

➮

AIntro



This program was written by Scott A.Basinger, Siddarayappa Bikkannavar,David Cohen, Joseph J. Green, John Lou,Catherine Ohara, David Redding, andFang Shi of Caltech for NASA’s Jet

Propulsion Laborator y. For furtherinformation, access the TechnicalSupport Package (TSP) free on-line atwww.techbriefs.com/tsp under the Softwarecategory.


Predicting Boundary-Layer Transition on Space-Shuttle Re-EntryLangley Research Center, Hampton, Virginia

The BLT Prediction Tool (“BLT” signi-fies “Boundary Layer Transition”) is pro-vided as part of the Damage AssessmentTeam analysis package, which is utilizedfor analyzing local aerothermodynamicsenvironments of damaged or repairedspace-shuttle thermal protection tiles.Such analyses are helpful in decidingwhether to repair launch-induced damagebefore re-entering the terrestrial atmos-phere. Given inputs that include re-entrytrajectory and attitude parameters, airdensity, air temperature, and details ofeach damage or repair site, the BLTPrediction Tool calculates expected timesof laminar-to-turbulent transition onset ofthe boundary-layer flow during re-entry.(These times help to define the properaerothermodynamic environment to usein subsequent thermal and stress analysesof local structural components.)

The BLT Prediction Tool includes adatabase of computed boundary-layerparameters that cover a range of nominalre-entry trajectories and uses an interpo-lation program for estimating localboundary-layer properties during flightalong a specific trajectory. Boundary-layer-trans ition criteria used in the BLTPre diction Tool were developed fromground-based measurements to accountfor effects of both protuberances and cav-ities, and have been calibrated againstflight data. Version 1 of this BLT predic-tion tool was developed in time for thefirst Return-to-Flight mission STS-114.

This work was done by Scott Berry, TomHorvath, Ron Merski, Derek Liechty, FrankGreene, Karen Bibb, and Greg Buck ofLangley Research Center; Harris Hamiltonand Jim Weilmuenster, Contractors withLangley Research Center; Chuck Campbell,Stan Bouslog, Ben Kirk, Garry Bourland,Amy Cassady, and Brian Anderson ofJohnson Space Center; Dan Reda and JamesReuther of Ames Research Center; GerryKinder, Dennis Chao, Jay Hyatt, MariaBarnwell, and K. C. Wang of The BoeingCo.; and Steve Schneider of PurdueUniversity. For more information, contact theLangley Innovative Partnerships office at(757) 864-4015. LAR-17337-1

Critical Elements for development of new BLT tool for on-orbit assessments.

Space Shuttle Model in wind tunnel tests.

Cov ToC + – ➭

➮

AIntro




A computer program performs calcu-lations to calibrate a quadrupole massspectrometer in an instrumentation sys-tem for identifying trace amounts oforganic chemicals in air. In the opera-tion of the mass spectrometer, the mass-to-charge ratio (m/z) of ions beingcounted at a given instant of time is afunction of the instantaneous value of arepeating ramp voltage waveformapplied to electrodes. The count rate asa function of time can be converted toan m/z spectrum (equivalent to a massspectrum for singly charged ions), pro-

vided that a calibration of m/z is avail-able.

The present computer program canperform the calibration in either orboth of two ways: (1) Following a data-based approach, it can utilize the count-rate peaks and the times thereof meas-ured when fed with air containingknown organic compounds. (2) It canutilize a theoretical proportionalitybetween the instantaneous m/z and theinstantaneous value of an oscillatingapplied voltage. The program can alsoestimate the error of the calibration per-

formed by the data-based approach. Ifcalibrations are performed in both ways,then the results can be compared toobtain further estimates of errors.

This program was written by SeungwonLee of Caltech for NASA’s Jet PropulsionLaboratory. For more information, down-load the Technical Support Package (freewhite paper) at www.techbriefs.com/tspunder the Software category.


Simulating the Gradually Deteriorating Performance of an RTGNASA’s Jet Propulsion Laboratory, Pasadena, California

Degra (now in version 3) is a com-puter program that simulates the per-formance of a radioisotope thermo-electric generator (RTG) over its life-time. Degra is provided with a graphi-cal user interface that is used to editinput parameters that describe the ini-tial state of the RTG and the time-vary-ing loads and environment to which itwill be exposed. Performance is com-puted by modeling the flows of heatfrom the radioactive source andthrough the thermocouples, also allow-ing for losses, to determine the tem-perature drop across the thermocou-

ples. This temperature drop is used todetermine the open-circuit voltage,electrical resistance, and thermal con-ductance of the thermocouples.Output power can then be computedby relating the open-circuit voltage andthe electrical resistance of the thermo-couples to a specified time-varying loadvoltage.

Degra accounts for the gradual dete-rioration of performance attributableprimarily to decay of the radioactivesource and secondarily to gradual dete-rioration of the thermoelectric materi-al. To provide guidance to an RTG

designer, given a minimum of input,Degra computes the dimensions, mass-es, and thermal conductances of impor-tant internal structures as well as theoverall external dimensions and totalmass.

This program was written by Eric G. Wood,Richard C. Ewell, Jagdish Patel, David R.Hanks, Juan A. Lozano, G. Jeffrey Snyder,and Larry Noon of Caltech for NASA’s JetPropulsion Laboratory.


2D/3D Synthetic Vision Navigation DisplayLangley Research Center, Hampton, Virginia

Flight-deck display software wasdesigned and developed at NASALangley Research Center to providetwo-dimensional (2D) and three-dimen-sional (3D) terrain, obstacle, and flight-path perspectives on a single navigationdisplay. The objective was to optimizethe presentation of synthetic vision (SV)system technology that permits pilots toview multiple perspectives of flight-deckdisplay symbology and 3D terrain infor-mation. Research was conducted to eval-uate the efficacy of the concept. Theconcept has numerous unique imple-mentation features that would permitenhanced operational concepts and

efficiencies in both current and futureaircraft.

One innovative feature, shown in thefigure, was the ability of the flight crewto select among several modes that pre -sent a dynamic 3D perspective of air-craft within the flight environment.The study focus was to uncover thedevelopments and benefits of using the2D and 3D exocentric SV informationwith regard to primary flight displays(PFDs) and navigational displays(NDs) for reducing accidents and dam-age for commercial aircraft. The inves-tigated technologies aim toward elimi-nating low visibility conditions as a

causal factor in civil aircraft accidents,while replicating the operational bene-fits of clear-day flight operations,regardless of actual outside visibilityconditions. The concepts also form thebasis of revolutionary electronic flightbag applications that utilize these tech-nological enhancements.

The results showed that SV on thePFD was pivotal for pilot use in terrainavoidance and situation awareness,while SV terrain on the 2D co-planarnavigational display was not found toprovide much benefit. However, pilotsnoted that the 3D exocentric display ofsynthetic terrain, with key implementa-

Calculations for Calibration of a Mass Spectrometer NASA’s Jet Propulsion Laboratory, Pasadena, California

Physical Sciences

Cov ToC + – ➭

➮

AIntro




Automated Camera Array Fine Calibration NASA’s Jet Propulsion Laboratory, Pasadena, California

tion features, added significantly toflight-crew situation awareness and sub-stantially enhanced the pilot’s ability todetect and avoid controlled-flight-into-terrain situations.

Conclusions reached indicate that SVdepicted on PFD is essential for terrainawareness. The situational awareness rat-ings for the SV PFD were largely due tothe egocentric view that gave pilots an

immersed sense of terrain around them.Pilot awareness and the capability foravoiding hazardous conditions were sig-nificantly enhanced with the addition of3D exocentric navigation display modesthat allowed for a greater field-of-regardto confirm the presence of hazardsalong their planned routing. The com-bination of SV primary flight and navi-gation display concepts allowed pilots to

make the best and quickest decisionsregarding safety of their aircraft.

This work was done by Lawrence J. PrinzelIII, Lynda J. Kramer, J.J. Arthur III, andRandall E. Bailey of Langley Research Centerand Jason L. Sweeters of NCI InformationSystems, Inc. For more information, downloadthe Technical Support Package (free whitepaper) at www.techbriefs.com/tsp under theSoftware category. LAR-17354

Using aerial imagery, the JPLFineCalibration (JPL FineCal) softwareautomatically tunes a set of existingCAHVOR camera models for an array ofcameras. The software finds matching fea-tures in the overlap region betweenimages from adjacent cameras, and usesthese features to refine the camera mod-els. It is not necessary to take specialimagery of a known target and no survey-ing is required.

JPL FineCal was developed for usewith an aerial, persistent surveillanceplatform. Synchronized images from anarray of cameras are captured andstitched together into a single, very high-resolution image that is projected ontoan elevation map of the ground. A GUI

(graphical user interface) tool allows theuser to play a movie of any part of theimaged surface from any perspective.

JPL FineCal requires, as input, a set ofCAHVOR camera models for the cam-era array. These models are typicallydeveloped on the ground using a cali-bration procedure requiring a knowntarget at a short distance. JPL FineCalcorrects the inaccuracy of the cameramodel extrinsic parameters resultingfrom the short target distance by usingimagery, taken during flight, at an effec-tive distance of infinity. It also makessmall improvements to the intrinsicparameters.

JPL FineCal is an automated processthat does not require the use of any spe-

cial targets, and which may be appliedduring normal flight operations. Thus, itmakes it simple to retune the cameramodels to correct for small misalignmentsthat occur due to changes in aperture set-tings, vibration, or thermal changes.

This work was done by Daniel Clouse,Curtis Padgett, Adnan Ansar, and YangCheng of Caltech for NASA’s Jet PropulsionLaboratory. For more information, down-load the Technical Support Package (freewhite paper) at www.techbriefs.com/tspunder the Software category.

The software used in this innovation isavailable for commercial licensing. Please con-tact Karina Edmonds of the CaliforniaInstitute of Technology at (626) 395-2322.Refer to NPO-45715.

Dynamic 3D Exocentric Navigation Display: Examples show several available display modes selected by pilots that dynamically present an immersed mov-ing 3D perspective of the aircraft in relation to terrain, flight path, and aviation hazards (represented here as static images).

Cov ToC + – ➭

➮

AIntro





Multichannel Networked Phasemeter Readout and Analysis NASA’s Jet Propulsion Laboratory, Pasadena, California

Netmeter software reads a data streamfrom up to 250 networked phasemeters,synchronizes the data, saves the reduceddata to disk (after applying a low-pass fil-ter), and provides a Web server interfacefor remote control. Unlike olderphasemeter software that requires a spe-cial, real-time operating system, this pro-gram can run on any general-purposecomputer. It needs about five percent ofthe CPU (central processing unit) toprocess 20 channels because it addsbuilt-in data logging and network-basedGUIs (graphical user interfaces) that are

implemented in Scalable VectorGraphics (SVG).

Netmeter runs on Linux andWindows. It displays the instantaneousdisplacements measured by severalphasemeters at a user-selectable rate, upto 1 kHz. The program monitors themeasure and reference channel fre-quencies. For ease of use, levels of statusin Netmeter are color coded: green fornormal operation, yellow for networkerrors, and red for optical misalignmentproblems. Netmeter includes user-selec-table filters up to 4 k samples, and user-

selectable averaging windows (after fil-tering). Before filtering, the programsaves raw data to disk using a burst-writetechnique.

This work was done by Shanti Rao ofCaltech for NASA’s Jet Propulsion Laboratory.For more information, download theTechnical Support Package (free whitepaper) at www.techbriefs.com/tsp under theSoftware category.


MISR Instrument Data Visualization NASA’s Jet Propulsion Laboratory, Pasadena, California

Platform for Postprocessing Waveform-Based NDEJohn H. Glenn Research Center, Cleveland, Ohio

The MISR Interactive eXplorer(MINX) software functions both as ageneral-purpose tool to visualizeMultiangle Imaging SpectroRadiometer(MISR) instrument data, and as a spe-cialized tool to analyze properties ofsmoke, dust, and volcanic plumes. Itincludes high-level options to createmap views of MISR orbit locations;scrollable, single-camera RGB (red-green-blue) images of MISR level 1B2(L1B2) radiance data; and animations ofthe nine MISR camera images that pro-vide a 3D perspective of the scenes thatMISR has acquired.

The plume height capability providesan accurate estimate of the injectionheight of plumes that is needed by airquality and climate modelers. MISR pro-vides global high-quality stereo heightinformation, and this program uses thatinformation to perform detailed heightretrievals of aerosol plumes. Users can

interactively digitize smoke, dust, or vol-canic plumes and automatically retrieveheights and winds, and can also archiveMISR albedos and aerosol properties, aswell as fire power and brightness temper-atures associated with smoke plumesderived from Moderate ResolutionImaging Spectroradiometer (MODIS)data.

Some of the specialized options inMINX enable the user to do other tasks.Users can display plots of top-of-atmosphere bidirectional reflectancefactors (BRFs) versus camera-angle forselected pixels. Images and animationscan be saved to disk in various formats.Also, users can apply a geometric regis-tration correction to warp cameraimages when the standard processingcorrection is inadequate. It is possible todifference the images of two MISR orbitsthat share a path (identical groundtrack), as well as to construct pseudo-

color images by assigning different com-binations of MISR channels (angle orspectral band) to the RGB display chan-nels.

This software is an interactive applica-tion written in IDL and compiled into anIDL Virtual Machine (VM) “.sav” file.

This work was done by David Nelson ofColumbus Technologies and Services Inc.;Michael Garay of Raytheon; David Diner,Charles Thompson, Jeffrey Hall, BrianRheingans, and Dominic Mazzoni of Caltechfor NASA’s Jet Propulsion Laboratory. Formore information, download the TechnicalSupport Package (free white paper) atwww.techbriefs.com/tsp under the Softwarecategory.


Taking advantage of the similaritiesthat exist among all waveform-basednon-destructive evaluation (NDE) meth-ods, a common software platform hasbeen developed containing multiple-sig-nal and image-processing techniques forwaveforms and images. The NASA NDESignal and Image Processing software

has been developed using the latest ver-sions of LabVIEW, and its associatedAdvanced Signal Processing and VisionToolkits. The software is useable on a PCwith Windows XP and Windows Vista.

The software has been designed with acommercial grade interface in whichtwo main windows, Waveform Window

and Image Window, are displayed if theuser chooses a “waveform file” to display.Within these two main windows (see fig-ure), most actions are chosen throughlogically conceived run-time menus. TheWaveform Window has plots for both theraw time-domain waves and their fre-quency-domain transformations (fast

Physical Sciences

Cov ToC + – ➭

➮

AIntro





Fourier transform and power spectraldensity). The Image Window shows theC-scan image formed from informationof the time-domain waveform (such aspeak amplitude) or its frequency-domain transformation at each scanlocation. The user also has the ability toopen an image, or series of images, or asimple set of X-Y paired data set in textformat. Each of the Waveform andImage Windows contains menus fromwhich to perform many user actions.

An option exists to use raw wavesobtained directly from scan, or wavesafter deconvolution if system waveresponse is provided. Two types ofdeconvolution, time-based subtractionor inverse-filter, can be performed toarrive at a deconvolved wave set.Additionally, the menu on the WaveformWindow allows preprocessing of wave-forms prior to image formation, scalingand display of waveforms, formation ofdifferent types of images (includingnon-standard types such as velocity), gat-ing of portions of waves prior to imageformation, and several other miscella-neous and specialized operations.

The menu available on the ImageWindow allows many further image pro-cessing and analysis operations, some ofwhich are found in commercially-avail-able image-processing software pro-grams (such as Adobe Photoshop), andsome that are not (removing outliers, B-

scan information, region-of-interestanalysis, line profiles, and precision fea-ture measurements).

This work was done by Don Roth for GlennResearch Center. For more information, down-load the Technical Support Package (freewhite paper) at www.techbriefs.com/tspunder the Software category.

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, InnovativePartnerships Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road,Cleveland, Ohio 44135. Refer to LEW-18261-1.

Graphical-User-Interface Windows are shown for NASA NDE wave and image processor.

WaveformWindow

ImageWindow

System for Continuous Delivery of MODIS Imagery to InternetMapping Applications NASA’s Jet Propulsion Laboratory, Pasadena, California

This software represents a complete,unsupervised processing chain thatgenerates a continuously updatingglobal image of the Earth from themost recent available MODIS Level 1Bscenes.

The software constantly updates aglobal image of the Earth at 250 m perpixel. It uses an event-driven schedulerto manage asynchronous image genera-tion tasks on a cluster of computers. Theoutput composite image is tightly inte-grated in the JPL OnEarth WMS server,which offers direct access to the globalimage to any Web Mapping Service(WMS) compatible client, and also sup-ports KML generation for Google Earth.The resulting Earth image composite ispermanently available as an Internetservice to WMS compatible mapping

applications. This application can han-dle the throughput of the MODIS satel-lite, processing more than 80GB of Level1B input data each day.

There are two main components tothis software package: DailyHarvest andDailyPlanet. The first component is ascene harvester that manages a localcopy of available MODIS scenes for thepast few days. When active, theDailyHarvest module checks the currentstate of the MODIS source repository(LAADS) against the local state. Anynew scene is downloaded, and the localcopy state is altered to reflect the avail-ability of the new scene. The secondcomponent, DailyPlanet, is then madeaware of the change in the archive state.Once the remote and local scenes aresynchronized, the DailyHarvest module

reschedules itself, effectively runningevery few minutes.

The DailyPlanet module functions asan event-driven scheduler that managesthe scene transitions from raw scenes tothe global composite. There are threeseparate scene queues: a raw scenequeue, a confirmed scene queue, and animage fragment queue. Each scenemakes the transition to the next queuebased on the result of an externalprocess. The transition from raw-to-confirmed is done by extracting thescene metadata and applying a suitabilitytest, confirming that the scene is not anight scene, and that the latitude rangeis within the S72-N72. The transitionfrom the confirmed scene queue to theimage fragment queue is the result of thesuccessful completion of an external task

Cov ToC + – ➭

➮

AIntro




that produces a visual image in geo-graphical coordinates from a MODISLevel1B scene. This processing is doneusing the HDFLook MODIS softwarepackage, running on a remote computer.

The final task is the integration of theimage fragments into the global compos-

ite, handled by a custom written externaltask. If an error is detected during any ofthe transitions, the error is logged andthe scene is dropped from the processingqueues. Most of the remote processingresources can be configured and areused in parallel for greater efficiency.

This work was done by Lucian Plesea ofCaltech for NASA’s Jet Propulsion Laboratory.


Automatic Rock Detection and Mapping from HiRISE ImageryNASA’s Jet Propulsion Laboratory, Pasadena, California

This system includes a C-code soft-ware program and a set of MATLABsoftware tools for statistical analysis androck distribution mapping. The majorfunctions include rock detection androck detection validation. The rockdetection code has been evolved into aproduction tool that can be used byengineers and geologists with minortraining.

The software takes as an input animage of a scene containing rocks andproduces as output a description of therock population and associated statistics.Each rock is described in terms of loca-tion, dimensions, and confidence ofdetection. The input parameters arethe image resolution (ground sampling

distance, or the size of a pixel in cen-timeters), the Sun incidence andazimuth angles for analysis of the shad-ows cast by rocks to derive individualrock models, and a parameter than canbe adjusted to accommodate variationsin image contrast.

The software is able to process verylarge reconnaissance imagery using astandard desktop computer by automat-ically processing image blocks and col-lecting all output in a single rock popu-lation description file (RPDF). Pro -cessing time is in the order of minutesfor nominal HiRISE images covering6×12 km areas at 30 cm/pixel.

The test option allows small portionsof the large images to be selected and

processed. Alternatively, a specificimage window can be processed byindicating its coordinates and size. Inthis mode, visual results (detectionsoverlaid on the images) are providedin addition to the rock population file(RPDF). This option is useful to quick-ly allow verification of parameter set-tings, and the quality of the detectionresults.

This work was done by Andres Huertas,Douglas S. Adams, and Yang Cheng ofCaltech for NASA’s Jet Propulsion Laboratory.


Parallel Computing for the Computed-Tomography ImagingSpectrometerNASA’s Jet Propulsion Laboratory, Pasadena, California

This software computes the tomograph-ic reconstruction of spatial-spectral datafrom raw detector images of the Com -puted-Tomography Imaging Spec trom -eter (CTIS), which enables transient-level,multi-spectral imaging by capturing spatialand spectral information in a single snap-shot. The CTIS can be used for surveyingplanetary landscapes through spectralimaging. It can also be used for battlefieldsurveillance and the spectral imaging oflive tissues for disease detection.

A Message Passing Interface Library(MPI) is used to parallelize the origi-nal serial version of the code withoutmodifying its initial structure. By par-allelizing the code, a speedup of up to20 is reached by using 32 processors.The software does not use any third-party libraries that require licenses. Itis written in Fortran and MPI, and thestorage of matrix elements is effi-cient, thus reducing memory require-ments.

This work was done by Seungwon Lee ofCaltech for NASA’s Jet Propulsion Laboratory.For more information, download theTechnical Support Package (free whitepaper) at www.techbriefs.com/tsp under theSoftware category.


Processing LiDAR Data To Predict Natural HazardsStennis Space Center, Mississippi

ELF-Base and ELF-Hazards (wherein“ELF” signifies “Extract LiDAR Features”and “LiDAR” signifies “light detectionand ranging”) are developmental soft-

ware modules for processing remote-sensing LiDAR data to identify past nat-ural hazards (principally, landslides)and predict future ones. ELF-Base

processes raw LiDAR data, includingLiDAR intensity data that are oftenignored in other software, to create digi-tal terrain models (DTMs) and digital

Physical Sciences

Cov ToC + – ➭

➮

AIntro




feature models (DFMs) with sub-meteraccuracy.

ELF-Hazards fuses raw LiDAR data,data from multispectral and hyperspectraloptical images, and DTMs and DFMs gen-erated by ELF-Base to generate hazardrisk maps. Advanced algorithms in thesesoftware modules include line-enhance-ment and edge-detection algorithms, sur-face-characterization algorithms, andalgorithms that implement innovativedata-fusion techniques. The line-extrac-tion and edge-detection algorithmsenable users to locate such features asfaults and landslide headwall scarps.

Also implemented in this software areimproved methodologies for identifica-tion and mapping of past landslideevents by use of (1) accurate, ELF-derived surface characterizations and(2) three LiDAR/optical-data-fusiontechniques: post-classification datafusion, maximum-likelihood estimationmodeling, and hierarchical within-classdiscrimination. This software is expectedto enable faster, more accurate forecast-ing of natural hazards than has previous-ly been possible.

This program was written by IanFairweather and Robert Crabtree of

HyPerspectives Inc. and Stacey Hager ofYellowstone Ecological Research Center forStennis Space Center.

Inquiries concerning rights for its commer-cial use should be addressed to:

Robert CrabtreeHyPerspectives Inc.2048 Analysis Drive Suite CBozeman, MT 59718E-mail: [email protected] No.: (406) 556-9880Refer to SSC-00279, volume and number

of this NASA Tech Briefs issue, and thepage number.

Information Sciences

Rock Segmentation Through Edge RegroupingNASA’s Jet Propulsion Laboratory, Pasadena, California

Rockster is an algorithm that automati-cally identifies the locations and bound-aries of rocks imaged by the rover hazardcameras (hazcams), navigation cameras(navcams), or panoramic cameras (pan-cams). The software uses edge detectionand edge regrouping to identify closedcontours that separate the rocks from thebackground (see figure). The algorithmhas applications both in ground-baseddata analysis, for example, to examine

large quantities of images returned by theMars Exploration Rovers, and in onboard(on-rover) opportunistic science applica-tions such as construction of rock mapsduring traverse, identification of unusualor otherwise high-value science targetsthat warrant additional investigation, anddetection of certain types of geologic con-tact zones.

The software uses gray-level intensitygradients to identify raw contours; these

raw contours are then split into shorter,low-curvature fragments. New fragmentsare created where necessary to bridgeareas of poor gradient information orpoor image quality. The algorithm uses aflooding step to regroup the variousfragments into closed contours. Thealgorithm is very fast with the C imple-mentation able to process (768×1024)images containing hundreds to thou-sands of rocks in approximately one sec-ond on a desktop workstation.

The algorithm is particularly efficientat quickly detecting small- to medium-sized rocks with sufficient contrast (posi-tive or negative) relative to the back-ground. Full quantitative performancecomparisons are not yet available; how-ever, preliminary tests show thatRockster appears to detect a significant-ly larger fraction of rocks present in ascene (higher recall) than previous rockdetection schemes, while maintaining ahigh precision rate (objects identified asrocks, truly are rocks).

Rockster has been integrated success-fully into a number of recent, high-leveldemonstrations, including the SOOPS(Science Operations on PlanetarySurfaces) demo, which used a rockexploration scenario to let scientists gainhands-on experience with anautonomous science capability in a sim-ulated environment, and live exercisesAn Overview of the Rockster rock segmentation algorithm.

Rockster Overview

Sky — GroundSegmentation

Edge Detectionand Linking

Salient Point;Splitting; T-Junction

and Gap Filling

Incoming Image

Initial Contours

BackgroundFlooding

Rock mask

Final Result

Cov ToC + – ➭

➮

AIntro



of the OASIS (Onboard AutonomousScience Investigation System)/CLARAty(Coupled Layer Architecture forRobotic Autonomy) software which werecarried out in real-time in the JPL Mars

Yard onboard the FIDO Rover (a closerelative of the twin Mars ExplorationRovers).

This program was written by Michael Burl ofCaltech for NASA’s Jet Propulsion Laboratory.


DSN Array SimulatorNASA’s Jet Propulsion Laboratory, Pasadena, California

The DSN Array Simulator (wherein“DSN” signifies NASA’s Deep SpaceNetwork) is an updated version of soft-ware previously denoted the DSNReceive Array Technology AssessmentSimulation. This software (see figure) isused for computational modeling of aproposed DSN facility comprising user-defined arrays of antennas and transmit-ting and receiving equipment formicrowave communication with space-craft on interplanetary missions. Thesimulation includes variations in space-craft tracked and communicationdemand changes for up to severaldecades of future operation. Such mod-eling is performed to estimate facilityperformance, evaluate requirements

that govern facility design, and evaluateproposed improvements in hardwareand/or software.

The updated version of this softwareaffords enhanced capability for charac-terizing facility performance againstuser-defined mission sets. The softwareincludes a Monte Carlo simulation com-ponent that enables rapid generation ofkey mission-set metrics (e.g., numbers oflinks, data rates, and date volumes), andstatistical distributions thereof as func-tions of time. The updated version alsooffers expanded capability for mixed-asset network modeling — for example,for running scenarios that involve user-definable mixtures of antennas havingdifferent diameters (in contradistinc -

tion to a fixed number of antennas having the same fixed diameter). The im -proved version also affords greater simu-lation fidelity, sufficient for validation bycomparison with actual DSN operationsand analytically predictable perform-ance metrics.

This program was written by RaffiTikidjian and Ryan Mackey of Caltech forNASA’s Jet Propulsion Laboratory. For moreinformation, download the TechnicalSupport Package (free white paper) atwww.techbriefs.com/tsp under the Softwarecategory.


DSN Array Simulator Overview

Block Diagram of Simulator Inputs & Outputs

Spacecraft Profile

• Mission Start & End Periods• Downlink & Uplink Data Rates• Tracking Frequency & Duration• Mission Probabilities• Destination• Etc …

DSN SimulatorSoftware

Simulation Period• Number of Cycles• Excel Output Template• Enable / Disable WeatherEffects, Visibility Models, andothers• Etc…

DSN NetworkConfiguration

• Antenna Type Definitions• Maintenance Schedules• Fault Descriptions• Crew Specification• Etc …

Policy AssumptionsManagement

• Shared Beam Operation• How manysimultaneous tracks?

• Antenna Operation• How much time toinitialize a track?

• Etc …

ConfigurableMetrics

Examples:• Peak Antenna Usage• Realized Downlinks &Uplinks• Realized Downlink &Uplink Data Rates• Potential Downlinks &Uplinks• Potential Downlink &Uplink Data Rates• Overall link metrics(achieved vs. rejections)

• Grouping by• antenna types• frequency bands• complex• link status

• Enveloping of datacontaining min. and max.variances

VisibilityModels

Graphical User Interface

WeatherModels

PolicyManager

ResourceManager

MetricsFramework

RuntimeEngine


Cov ToC + – ➭

➮

AIntro




Estimating Software-Development Costs With Greater AccuracyNASA’s Jet Propulsion Laboratory, Pasadena, California

Visual Target Tracking on the Mars Exploration RoversNASA’s Jet Propulsion Laboratory, Pasadena, California

COCOMOST is a computer programfor use in estimating software develop-ment costs. The goal in the developmentof COCOMOST was to increase estima-tion accuracy in three ways: (1) develop aset of sensitivity software tools that returnnot only estimates of costs but also the esti-mation error; (2) using the sensitivity soft-ware tools, precisely define the quantitiesof data needed to adequately tune costestimation models; and (3) build a reposi-tory of software-cost-estimation informa-tion that NASA managers can retrieve toimprove the estimates of costs of develop-ing software for their project (see figure).

COCOMOST implements a method-ology, called “2cee,” in which a uniquecombination of well-known pre-existingdata-mining and software-development-effort-estimation techniques are used toincrease the accuracy of estimates.COCOMOST utilizes multiple models toanalyze historical data pertaining to soft-ware-development projects and per-forms an exhaustive data-mining searchover the space of model parameters toimprove the performances of effort-esti-mation models. Thus, it is possible toboth calibrate and generate estimates atthe same time. COCOMOST is written

in the C language for execution in theUNIX operating system.

This program was written by Tim Menziesand Dan Baker of West Virginia Universityand Jairus Hihn and Karen Lum of Caltechfor NASA’s Jet Propulsion Laboratory. Formore information, download the TechnicalSupport Package (free white paper) atwww.techbriefs.com/tsp under the Softwarecategory.


Example Model Output

• Main Output: Cumulativedistributionfunction (S-curve)

• Recommend a range of 50% to 70% probability based on experience with past projects

Visual target tracking (VTT) softwarehas been incorporated into Release 9.2 ofthe Mars Exploration Rover (MER) flightsoftware, now running aboard the roversSpirit and Opportunity. In the VTT oper-

ation (see figure), the rover is driven inshort steps between stops and, at eachstop, still images are acquired by activelyaimed navigation cameras (navcams) ona mast on the rover (see artistic rendi-

tion). The VTT software processes thedigitized navcam images so as to track atarget reliably and to make it possible toapproach the target accurately to within afew centimeters over a 10-m traverse.

Cov ToC + – ➭

➮

AIntro




Functional Flow of VTT integrated into MER flight software. VTT module functions (square boxes) are additions to the existing MER flight software (circles).

The operations on the digitizedimages include a normalized cross-corre-lation algorithm along with template-image-magnification and template-image-roll-compensation algorithms.Each VTT update takes about 50 sec-onds. VTT has helped to make it possi-ble to approach a target over a 10-m tra-verse and place an instrument on thetarget during a single sol, whereas previ-ously, such approach and placementtook 3 sols. Alternatively, VTT can beused to simply image a target as therover passes it. VTT can be used in con-junction with any combination of blinddriving, autonomous navigation withhazard avoidance, and/or visual odome-try.

This program was written by Won Kim,Jeffrey Biesiadecki, and Khaled Ali of Caltechfor NASA’s Jet Propulsion Laboratory.


Visual Odometry Drive RoverHazard Avoidance Auto-Nav

VTT CameraPointing

VTT UpdateTarget

VTT SeedTarget

Capture Image End

Start

Artistic Rendition of MER.

Parametric-Studies and Data-Plotting Modules for the SOAPNASA’s Jet Propulsion Laboratory, Pasadena, California

“Parametric Studies” and “Data TablePlot View” are the names of softwaremodules in the Satellite Orbit AnalysisProgram (SOAP). Parametric Studiesenables parameterization of as many asthree satellite or ground-station attrib-utes across a range of values and com-putes the average, minimum, and maxi-

mum of a specified metric, the revisittime, or 21 other functions at each pointin the parameter space. This computa-tion produces a one-, two-, or three-dimensional table of data representingstatistical results across the parameterspace. Inasmuch as the output of a para-metric study in three dimensions can be

a very large data set, visualization is aparamount means of discovering trendsin the data (see figure).

Data Table Plot View enables visualiza-tion of the data table created byParametric Studies or by another datasource: this module quickly generates adisplay of the data in the form of a rotat-


Cov ToC + – ➭

➮

AIntro



Testing of Error-Correcting Sparse Permutation Channel CodesNASA’s Jet Propulsion Laboratory, Pasadena, California

able three-dimensional-appearing plot,making it unnecessary to load the SOAPoutput data into a separate plotting pro-gram. The rotatable three-dimensional-appearing plot makes it easy to deter-mine which points in the parameterspace are most desirable. Both modulesprovide intuitive user interfaces for easeof use.

This work was done by Robert Carnright,Adam Loverro, and Robert Oberto of Caltechand David Stodden and John Coggi of TheAerospace Corporation for NASA’s JetPropulsion Laboratory. For more information,download the Technical Support Package(free white paper) at www.techbriefs.com/tspunder the Software category.

This software is available for commerciallicensing. Please contact Karina Edmonds ofthe California Institute of Technology at(626) 395-2322. Refer to NPO-45059.Sample Output

A computer program performs MonteCarlo direct numerical simulations fortesting sparse permutation channelcodes, which offer strong error-correc-tion capabilities at high code rates andare considered especially suitable for stor-age of digital data in holographic and vol-ume memories. A word in a code of thistype is characterized by, among otherthings, a sparseness parameter (M) and afixed number (K) of 1 or “on” bits in achannel block length of N (see figure).

In a test, random user data words aregenerated and mapped into code words.Transmission of the words through anoisy channel is simulated by adding sim-ulated white Gaussian noise whose ampli-tude is determined by the signal-to-noiseratio. Detection of each word is per-formed via strict Maximum Likelihooddetection algorithm starting with the ini-tially detected codeword, which is pro-duced by sorting the resulting bit valuesand assigning a value of 1 to the highest

K of them and 0 to the rest. A newlydeveloped permutation sorting algo-rithm is further applied to determine themaximum weight valid code word. Themaximum likelihood valid sparse code-word is decoded, by means of an inverseof an enumerative scheme used in gen-erating code words, to obtain the origi-nal use data. From the results of this sim-ulation, block error rates (BERs) andother statistics that characterize the per-formance of the code are calculated.Sparse permutation channel codes satisfythe balanced channel coding constraintand, as determined from the directnumerical simulation, also simultaneous-ly provide high error correction with useof BER as low as 10–10 and less even forfairly large raw bit error rates. Channelcodes of sufficiently large block size (N >100) asymptotically approach the infor-mation theoretic limits for communica-tions channel capacity.

This program was written by Kirill V.Shcheglov and Sergei S. Orlov of Stanford andwas tested numerically by Hongtao Liu andSnezhana I. Abarzhi of Illinois Institute ofTechnology for NASA’s Jet PropulsionLaboratory. For more information, downloadthe Technical Support Package (free whitepaper) at www.techbriefs.com/tsp under theSoftware category.

This software is available for commerciallicensing. Please contact Karina Edmonds ofthe California Institute of Technology at (626)395-2322. Refer to NPO-45196.An Outline of the Code used for numerical simulation of the error-correction performance.

User data

The codeword is decoded using theinverse enumeration algorithm and ischecked for validity:

Constellation encoding is applied to userdata. The data is mapped into one sampleof a set of the sparseness M

The sparse data is encoded into a Hammingconstrained codeword having N bits, K ofwhich are 1’s using an enumeration algorithm

Noisy grayscale signal is detected

The first maximum likelihood decodingcorresponds to assigning 1’s to Kbrightest bits and 0’s to the rest

Valid Invalid

Bits are permuted in the order ofdecreasing likelihood. Next most likelyconstrained codeword is generated.

Here the noise causes two bit errors.

Decoded data from thesparse set is mappedback into the user data

Cov ToC + – ➭

➮

AIntro





Enhanced Reporting of Mars Exploration Rover TelemetryNASA’s Jet Propulsion Laboratory, Pasadena, California

Mars Exploration Rover EnhancedTelemetry Extraction and ReportingSystem (METERS) is software that gener-ates a human-readable representation ofthe state of the mobility and arm-relatedsystems of the Mars Exploration Rover(MER) vehicles on each Martian solar

day (sol). Data are received from theMER spacecraft in multiple streams hav-ing various formats including text mes-sages, sparsely-sampled engineeringquantities, images, and individual motor-command histories. Typically, only partsof this data generated on Mars are

received the same day they were created,so the summary report tools have to workwell even when data is missing. All infor-mation is grouped by type into easily-browsable Web pages (see Figure 1).

METERS is the first software to pro-vide an integrated view of the mobility

Figure 1. This top view of a METERS Auto-Generated Main Web Pageincludes a motion summary in plain text, graphical course plot, and a pull-down menu to select individual reports.

Figure 2. Another page includes Thumbnail Views of All Images, includingthumbnail-sized elevation data automatically extracted from stereo pairs ofimages.

A SPICE module for the SatelliteOrbit Analysis Program (SOAP) pre-cisely represents complex motionand maneuvers in an interactive, 3Danimated environment with supportfor user-defined quantitative outputs.(“SPICE” stands for Spacecraft,Planet, Instrument, Camera-matrix,and Events). This module enables theSOAP software to exploit NASA mis-sion ephemeris represented in theJPL Ancillary Information Facility(NAIF) SPICE formats. Ephemeristypes supported include position,velocity, and orientation for space-craft and planetary bodies includingthe Sun, planets, natural satellites,comets, and asteroids. Entire mis-sions can now be imported into SOAPfor 3D visualization, playback, andanalysis.

The SOAP analysis and display fea-tures can now leverage detailed missionfiles to offer the analyst both a numeri-cally correct and aesthetically pleasingcombination of results that can be variedto study many hypothetical scenarios.The software provides a modeling andsimulation environment that canencompass a broad variety of problemsusing orbital prediction. For example,ground coverage analysis, communica-tions analysis, power and thermal analy-sis, and 3D visualization that provide theuser with insight into complex geomet-ric relations are included.

The SOAP SPICE module allows dis-tributed science and engineering teamsto share common mission models ofknown pedigree, which greatly reducesduplication of effort and the potentialfor error. The use of the software spans

all phases of the space system lifecycle,from the study of future concepts tooperations and anomaly analysis. Itallows SOAP software to correctly posi-tion and orient all of the principal bod-ies of the Solar System within a singlesimulation session along with multiplespacecraft trajectories and the orienta-tion of mission payloads. In addition tothe 3D visualization, the user can definenumeric variables and x–y plots to quan-titatively assess metrics of interest.

This work was done by Robert Carnrightand Claude Hildebrand of Caltech andDavid Stodden and John Coggi of TheAerospace Corporation for NASA’s JetPropulsion Laboratory.


SPICE Module for the Satellite Orbit Analysis Program (SOAP)NASA’s Jet Propulsion Laboratory, Pasadena, California


Cov ToC + – ➭

➮

AIntro



Facilitating Analysis of Multiple Partial Data StreamsNASA’s Jet Propulsion Laboratory, Pasadena, California

Robotic Operations Automation:Mechanisms, Imaging, Navigation reportGeneration (ROAMING) is a set of com-puter programs that facilitates and accel-erates both tactical and strategic analysis oftime-sampled data — especially the dis-parate and often incomplete streams ofMars Explorer Rover (MER) telemetrydata described in the immediately preced-ing article. As used here, “tactical” refers tothe activities over a relatively short time(one Martian day in the original MERapplication) and “strategic” refers to alonger time (the entire multi-year MERmissions in the original application).

Prior to installation, ROAMINGmust be configured with the types ofdata of interest, and parsers must bemodified to understand the format ofthe input data (many example parsersare provided, including for generalCSV files). Thereafter, new data frommultiple disparate sources are auto-matically resampled into a single com-mon annotated spreadsheet stored in areadable space-separated format, andthese data can be processed or plottedat any time scale. Such processing orplotting makes it possible to study notonly the details of a particular activity

spanning only a fewseconds, but alsolonger-term trends.ROAMING makes itpossible to generatemission-wide plotsof multiple engi-neering quantities

[e.g., vehicle tilt as in Figure 1(a),motor current, numbers of images]that, heretofore could be found only inthousands of separate files.

ROAMING also supports automaticannotation of both images and graphs.In the MER application, labels given toterrain features by rover scientists andengineers are automatically plotted inall received images based on their asso-ciated camera models (see Figure 2),times measured in seconds are mappedto Mars local time, and command namesor arbitrary time-labeled events can beused to label engineering plots, as inFigure 1(b).

This work was done by Mark W. Maimoneand Robert R. Liebersbach of Caltech forNASA’s Jet Propulsion Laboratory.


Figure 1. Plots may include (a) multiple fields or (b) event-based colorcoding, and support interactive zooming. Figure 2. Images are automatically annotated with detailed drive information.

(a)

(b)

and arm operations even when data aremissing. METERS combines the datareceived in the various telemetry streamsduring an entire sol, making clear whathas occurred and annotating what dataare missing. METERS comprises a set ofsoftware tools (primarily in C, C++) andPerl language scripts for robustly com-bining these data into a concise, human-

readable format. Raw data products areconverted into Hypertext MarkupLanguage files, compatible with Web-browser software, that include thumbnailimages (see Figure 2), summaries ofmotions, and plots of engineering data.Automatically generated reports alsosummarize the classes of mobility andarm activities that occur during each sol.

This work was done by Mark W. Maimone,Jeffrey J. Biesiadecki, Robert R. Liebersbach,Joseph L. Carsten, and Chris Leger of Caltechfor NASA’s Jet Propulsion Laboratory.


Cov ToC + – ➭

➮

AIntro



Service-Oriented Architecture for NVO and TeraGridComputingNASA’s Jet Propulsion Laboratory, Pasadena, California

The National Virtual Observatory(NVO) Extensible Secure ScalableService Infrastructure (NESSSI) is aWeb service architecture and softwareframework (see figure) that enablesWeb-based astronomical data publish -ing and processing on grid computerssuch as the National Science Foun -dation’s TeraGrid. Charac ter istics ofthis architecture include the follow-ing: • Services are created, managed, and

upgraded by their developers, whoare trusted users of computing plat-forms on which the services aredeployed.

• Service jobs can be initiated by meansof Java or Python client programs runon a command line or with Web portals.

• Access is granted within a graduatedsecurity scheme in which the size of ajob that can be initiated depends onthe level of authentication of theuser. A “small” service request may be sub-

mitted anonymously. A “medium” requestmay be submitted with a “weak” certificateissued by the NVO or another certificateauthority not associated with an officialgrid computing organization like Tera -Grid or the Department of Energy (DOE).A “large” request must be accompanied by

a “strong” certificate issued by the TeraGridor DOE certificate authority. User certifi-cates are managed by the Clarens Grid-Enabled Web Services Framework(http://clarens.sourceforge.net/).

This work was done by Joseph Jacob, CraigMiller, Roy Williams, Conrad Steenberg, andMatthew Graham of Caltech for NASA’s Jet

Propulsion Laboratory. For more information,download the Technical Support Package(free white paper) at www.techbriefs.com/tspunder the Software category.


In this Model, the application server is Web-service container that accepts HTTP connections that maybe SOAP messages and may have a certificate attached.

trust

fetchcertificate

clienthttp form

SOAP http

SOAP httpportal

Community domain

application server

userService

userService

systemService

systemService

bus

CPU farm

node

node

nodemyspace

myspace

myspace

storage

Service-grid node

The MRO OLVM wrapper script soft-ware allows Mars ReconnaissanceOrbiter (MRO) sequence and space-craft engineers to rapidly simulate aspacecraft command product through atool that simulates the onboardsequence management software(OLVM). This script parses sequencefiles to determine the appropriate timeboundaries for the sequence, and con-structs the script file to be executed byOLVM to span the entirety of the desig-nated sequence. It then constructsscript files to be executed by OLVM,constructs the appropriate file directo-ries, populates these directories with

needed input files, initiates OLVM tosimulate the actual command productthat will be sent to the spacecraft, andcaptures the results of the simulationrun to an external file for later review.Additionally, the tool allows a user tomanually construct the script, ifdesired, and then execute the scriptwith a simple command line.

This work was done by Roy Gladden, ForestFisher, and Teerapat Khanampornpan ofCaltech for NASA’s Jet Propulsion Laboratory.

This software is available for commerciallicensing. Please contact Karina Edmonds ofthe California Institute of Technology at(626) 395-2322. Refer to NPO-45242. Mars Reconnaissance Orbiter on the launch pad.

Mars Reconnaissance Orbiter Wrapper ScriptNASA’s Jet Propulsion Laboratory, Pasadena, California


Cov ToC + – ➭

➮

AIntro



Multiphysics Resources

Introduction CDs Success Stories

Model GalleryHands-On Workshops


Cov ToC + – ➭

➮

AIntro




http://www.abpi.net/ntbpdfclicks/l.php?200809STBCVR3D

http://www.abpi.net/ntbpdfclicks/l.php?200809STBCVR3E

Hann

over

• Bo

ston

COMSOL Conference 2008, Oct 9-11Join the Leaders in Multiphysics Simulation

BOSTON HANNOVER TEL AVIV DHAHRAN SHANGHAI BEIJING SINGAPORE BUDAPEST KUALA LUMPUR MADRID TAIPEI TOKYOOCT 9-11 NOV 4-6 NOV 11 NOV 16 NOV 18 NOV 20 NOV 21 NOV 24 NOV 25 NOV 26 NOV 28 DEC 5

www.comsol.com/confcd

CONFERENCE SPONSORS

The COMSOL Conference CD contains slides from all the user presentations and select model files. More than 100,000 people have requested the cd so far.

nn

nnrr

o•

ooottososBB

CCOO

OMO SOLLOSMOO ConfeefeffenoC

e eenceee 2008, OO,8002ecnerre

Oct 9-11-9tcO 11

rran

nove

JooJ

ni tn he LeaderssredaeLehtnni

s in MMultiphysiccisyhpitluMnis

cs SimullumiSsc

.comsol.wwww.

COMShe TD containC

ethe user prreselect modethan 100,00equested trre

com/confcd

ence eronffeCSOLom allns slides frro

esentations andeorMel files.

ev00 people ha.the cd so farr.

NNEERREEFFNNOOCC

SSRROOSSNNOOPPSSEECCNN


Cov ToC + – ➭

➮

AIntro




Documents

Software ProgramsPrograms - WordPress.comThe COMSOL® enviro li i h i l i t nm ent empowers you to co l i i l i h i vr e nduct d li capabilities.CAD-impo rea st c tec n ca s mu interfaces