Advanced Design And Implementation Of A Control Architecture For . - NASA
:.,k. "" . to,a, In . g ,,cc, R boti s, and A ,n,at, AdvancedDesignace '-q-SAIRAS 99'r, Nuordw jK, The Nct h ld,d.and Implementationfor LongRangeAutonomousA. Martin-AlvarezPlanetaryt, S. Hayati,AutonomyJet Propulsionof a ControlArchitectureRoversR. Volpe,and Control -R. PetrasSectionLaboratory/ CaliforniaInstituteMail Stop 198-219of Technology4800 Oak Grove Drive,Pasadena,CA 91109e-mail:alex @ telerobotics.jpl.nasa.govTel: (818)354-4725Fax: (818)393-5007A,BSTRACTAn advanceddesignand implementationof a ControlArchitecturefor Long Range AutonomousPlanetaryRoversis presentedusing a hierarchicaltop-downtask decomposition,and the commonstructureof eachpresentedmanagers,designdesignis presentedas a commonintuitivearchitecturedesigners,of the son feedbackcontrolfor the designwhenprogrammers,in the classictheory.a rogrammingteampersonnel.of cyclicisis composedThedataofwholeprocessingand event-drivenreactionto achieveall the reasoningand behaviorsneeded.For this purpose,acommercialgraphicaltool is presentedthat includesthe mentionedcontrolcapabilities.Messagesqueues(AI)are usedfor olsystem runningin real time on top the JPL sthe sinergyIntelligenceand classiccontrolconceptsRange AutonomousPlanetaryRov, ersin having:m advancedControlIntelligenceshowa highlyRocky7 controllingbetweenArtificialArchitecturefor Long1. IntroductionA highly autonomousneeds appropriatelydelaysandtemporaryrover is desiredabstractedstatelossin a planetaexplorationmission.The human on groundand status feedbackin the telemetrydown link. Thus,of communicationsto Eartharenota problembecauseno realonlytimetimecontrolloops are closed via the up/downlink. This a lso reducesthe power consumptionof roversubsystemslike telecommunicationand makes more re. ources available for the actual locomotion.ControlneededcontrolArchitecturesto achievearchitecturesare key[7][10][11][12][19]paper is pionnerbut howeverin suggestingarchitecturesforRover [1][23].a highlyI Visitingscientistelementsto identify,define,andimplementall thethe requieredautonomy.Then from the last decade,definitionhave been of major interest in the research field of autonomouscontrolaspectsand designofmobile robotsvery little have been done for a high challengingscenario.Thistools and proceduresto design, implement,and maintaincontrolcomplexmobilerobots futureat JPLAleianth'M lrtln-Al; rez/ 1/25LongRangeAutonomousPlanetary.
SubmittedThisto: pacewhatneededare ofThecontrolbuildingis to buildandwherethe desiredhardware,or humanThis paperpresentsframeworks,of rogramming.in the designChapter2 describesfirstArchitectureIntegratedof ControlRover.rocky72. Definitionamongthem.Followingto supportorbiters,Control layermission,Rover.as a commonandengineers,[8] are usedfor suchframeworkfor theamongthemand ChapterLaterandControland DataChapter5 givesMarsControlobjectivesand StateRobotArchitecturefor a highlyDiagrams4 describesexperimentalControlof a LongArchitectureFlowRover,Roverthe enario.are giventhe OperationalStatetheare shownthe basicresultapproachwiththe JPLa high level task.whatfor a Spaceare the neededis the first phaseRovercontrolfunctionsand interactionsof the design.(ICA)we first definedealingrovers.controlthe Integratedwith severalThe majorarchitecturescooperativecharacteristicsof alltheControlelementsArchitecturesuch(ICA)as roboticarms,of ICA as as inputfor eachspaceis definedelement.controllingof controloverallthe communicationon top of all the elementa commonfor exampleoptimalof each typeSpaceinteractionsRangeputfor the FunctionalArchitectureapproach,the placementto achieve showsgeneralof a spacecommands ethe top-downthe a commonControlfor a olArchitectureexecutingand planetaryplanetaryrover.this architectureControllanders,functions.softwareandor on-An Operationaldesigners,Shellanto the control ones.ho. wto implementarchitecturea real-timeon-boardthe designarchitecturefunctionsdefiningis to buildAutonomousof Control(ICA)3 describesusingControlwhenCreating2.1 Integratedprovidesas the baselineplanetaryof the FunctionalThe FunctionalcontrolcontrolArchitecturefor MobileChapterof this architecturefor the Implementationmicro-rovertool to supporteditorsArchitectureof the overalis definedMarsautonomousfor the designControlControlArchitectures(MORCA)and realisticof theShellof all(eithersupportRangeRangephase.the FunctionalanScientificControlphasescenario.and operationLongArchitecturefunctionsin additionArchitecturedefiningof managers,The NetherlandsforSecondfunctionsfor a LongGraphicalmaintenanceinto the contextArtArchitectureRangeIn definingteamthemControlfor a spaceand an advancedfor thea designand maintenancei plementation,of turethem.controlall the controlprocedures,and the transitiona l Architecturealso defines extra operati.onalThe last design phase is to build an Implemeniation(software,controlall thesecontrol99", Noordwijk,a Functionalinteractionsdefiningto achievefor Space "i-SAIRASa completefirst phasefunctionsArchitectureare locatedand AutomationmissionThethat is decomposedglobalthe sequencefunctionsmissionof tasksof one spacelayerthe res.(plannedcontrolsTheor scheduled)the executioncommoninto taskof the / 2/25on differentspaceelementin order
Submittedto: ArtificialIntelligence,Robotics,and AutomationThe definitionof theseinteractionsrole in a spacemissionbased on planetarya better positionbetterpositionto carryamongout controla landerpositionsand attitudebetweenThe mainapportationof the Integratedthe functionalelements.controlcontrolto carrylanderarchitectureout landingsite or landingof a planetaryStatusMissionroverwhentime,TheplaysomeFor examplein realArchitectureNoordwijk,elementsare neededelements.functionsand landing99",in differentInteractionsof for SpaceIA1A1Task 1 JTaskIR1.R1Navig./Ielementsuplinkingeitherto this paperhas to be openLNaviIwithTask &IR2R1Navig.Savig.I' I Navig.Status/L,AlLI CornelsYLI-- A1R,& --ensor. II I ControInputsL 1 l ( utnuts('nc'extert)I -V"R,Sens. IInputs J-]I ]I .2.L.I Nawg.&lRI i Control,[, J. outputsV-VILR22.2 ObjectivesThe mainand Stateobjectives Generality:to achieveNavig.' INavtg.1 I OmOsTo be a ation:roboticinformation Unit2L: ToSystems.arms,architectures Robustness:environment,IOLII )I Pi otin ; "]" JoSens l.'npUtr]IArchitectureControlV'OutputsSens. l.),'nPutIIdControl'OutputsV-(ICA)anda betterunderstandingsystem,in orderandare exchangedsameframeworkallowassessmentas wellas forimplementationthe interactionlanders)or commandsin thetakingamongallamongparta eelements(severalidentifyingof uncertaintiesand limitedMartin-Alvarezimplementationsand performancesof cooperationactuators,Aleiandroare:of all the functionsto fulfillof a mission,possibleof the capabilitiesdifferentconfigurationsof h elements.theof the elementstaking part of a mission.To allow robustnessboth in the presenceand in the presencefor a planetaryto allowTo supportorbiters,Architectureof a controlunify Flexibility:ToimplementationpilotingIL--I I COntr l. r outputs1"]ControlTable1), independentof operationssoftware,and human intervention), Inter-Element.,0of the Artof a Functionalthe designspaceIIjI LR2Sens. l. lInpu't . ]I 'Fig. 2. IntegratedotherStatus0IR20 - ]-0 -----].]A,thatIcoVI/ sR2Navig.ilrelativeis to emphasizeto interact/iI A1is intrajectories.(ICA)statusllco; 's statusl, ,Statuslj ommanoare inroverI I CommandMissionR2Navigan importanta planetaryMission Layer,Netherlands/ controlcontrolthe knowledgein rover,sensors,of theand
Submitted to: Artificial Intelligence, Robotics, and Automation for Space "i-SAIRAS 99", Noordwijk, The Netherlands Quickness: Savings:To allowexecutionTo achieveMissionin real time,low costCommandsDefinitionParametersINSPECT AREABuild a topographicenvironment.SURVEY shell shield (Mars) or other rover spacecraft/rover area to survey RELAY DATARelay data from on-groundspaceelement(e.g. rover or lander)toground segment (e.g. Earth)Deploy and install instruments. location locationINSTALL INSTRUMENTmapofthe initial position surface to inspect final location location of groundof groundspace element segment deployment location type of instrument #COLLECT SAMPLECollect, retrieverock samples.RETRIEVE ITEMOff loadinglanderTableTherearewithouteacha greatnumbera commonmobilerobotapplicationmechanismsamongthem,of a betterArchitectureunderstandingfor a nrequiredcontrolarchitecturesArchitectures.In this organizationare presentat each resolution/abstractionlevelor .HybridarchitecturesThe centralexternalat enfunctionsis of greatandinterestwheninteractionfor the sakeFunctionalControlto handle the controlhaveschemebasedbeendefinedon a hierarchylevel to eitherto the upperor behaviorsof layers,generateactiondecisioncommandsone [10] [18] [19].architecturesof thein the architecturein the last ten years:is thatsystem.designa controlTheresincesystemis not anygoalsgoaldo not change[1 1] [12].architecturesa consequenceof a new architectureis donearchitectures[24] [26].idea of behavioralmanifestationstime.to the nextinformationreturnlocation [10][11][12][19].architecturespacecraftsof both.adjacentof planetaryterminologyin previousand a hybridfrom one problemdefinitionrequirebehavioral,of thedesignto lander location type of item vioralconsiststhefromandfor functionalprocessesto the lowerRoverforscenarioa commonitemsfunctionsapproachesPure Hierachicalmaking1. Soaceof controlandcargo sample location type of sample amount of sample rendez-vouspointsamples and assessmentof capabilities.Also, a completeLong Range AutonomousSpace Rover is mandatorythat was nevermainofof approachesdefinitionsoil oror analysetakingtryto combineof eitherhierarchicalan evolutionthe most relevantadvantagesAleiandro Martin-Alvarez/ 4/25andof behavioralbehavioralapproaches.architecturesof both approachesor the[2] [16].
dsPurehierarchicalarchitecturesaregoodto define clearly allrealtimeexecutionuncertaintieshavingand constraintsreal time executionwhento oursimplescenarios.advantages2.3 MObileToOurRobot[16] [18]For the designwe use the bilefunctionscontrolMORCAandsequencecomplexityof differenttasksand ineMarsRovertheinbut fail bothall of themScientificfromand hierarchicalarchitecturedefinitionare thewereusedproperlyinall theworkingin a very(MORCA)[2] wasapproachRangerobotcontrolandof commandspiloting,commandMobile Robot Space[12]TheRover,in the hierarchicalof the problemon navigation,in the following[11]are not in conflict.Autonomousfollowedof the commandsto a subsequent[10]or in preparation.decompositionroverthe literatureapproachesfor a Longof a complexor mobileArchitectureapproachesengineeringas it is shownControlare used in executionhighest level missioncommandsvia commandsto individualwheel control commands.TheHybridbut howeverallare goodreasoningarchitectureRobot[2] [5]. Thisupdatecorrespondbut fail inwitharchitectureshigh levelscenario.doneRangewell-knownthat behavioralan enario.task decompositionunderstanding,evenPurebeenfor a LongmappingcontrolArchitecturehavea hybridArchitectureprovedof a newControlis to defineframework,is whetherapplication.designsthe controlreasoningis high and in a veryseveralobjectives,[19]. MORCAlevelwith the environmentexplorationall theseo ly differenceRobotgoalControlas a generalhighand AI techniquesroveraccomplishdefinedandof controlplanetaryrapidlyrequiredscenariomuchof a real roverreactingthe intelligenceanswercomplextoothem,Thehas beenmotionidentified,levelallowslevelsto the mobilewheelMobileloweraof ritywith human commandsHigh level of messages between two people(the boss to his employee)Degree 6(Mission Comds)a) InspectareaDegree 5(NavigationCommands)a) Go to a locationCommandsfrom a person, in an unknowncity to a taxi driver to go to a specificaddressDegree 4(Piloting Comds)a) Stay in a direction until eventb) Follow/Reachan object untileventCommandsDegree 3(TrajectoryCommands)a)b)c)e)Driver commandsDegree 2(DeviceControlCmds)a) Steeringb) SpeedDegree 1( Control Outputs)a) 5 Volts to Motor 1b) Switch on a brakeb) Collect sampleBorder/FollowobjectGo straight objectStay in knows how to gocarbydriverantoreachoccupantanwhoto a learnerCar driver control actionsAngle2. Mobileto aElectricalmusclesRobotCommandAleiandro Martin-Alvarez / 5/25Signal from the nervouslevels.systemto
Submitted to: Artificial Intelligence, Robotics, and Automation for Space "i-SAIRAS 99", Noordwijk, The NetherlandsTable2 showstransformeda hierarchyintoa setin mobilerobotof commandscommandsare performedmentionedhierarchycommandsof degreeby a set of functionsin mobileroboti-l.whereThecontainedcommands,all commandsdecompositionin a controlour approachstructure,presence),MORCAwhereis structuredfunctional Forwardcommand Nominalmodels")based(FC).ResponsibleControlof functionalon the conceptlayersof feedbackfor activityarchitecture.of suchBecauseof theof MObileRobotbased on differentlayers. Thethe more complicatedare theeachcontroldecomposition,i can becontrollayeris structuredinto three[7] [ ackbasedfc rnulation(NF).Functionson the actual,of controlled Non-Nominalanalysinginto a hierarchybranchesthefor the definitionControl Architecture(MORCA)also follows a hierarchicalmore the layers are able to work autonomously(no humantasks which the rover achieve itself.parallelof degreeandrefinementbut xpected,careIt containsevolutionof the correctfunctionsactual and allowablestates in both the FC andgenerationof directivesand constraintsfor ngNFfunctions,NominalFeedback1. Mobileknowledgeof the processandRobotControlAleiandroof this layer,for the monitoringNonFeedbackFigureof a priori("worldconsequentlyof the FC.takessituations.andArchitecture.Martin-Alvarez/ 6/25detectingof discrepanciesdiagnosisof theirandbetweenorigins,and
Submitted to: Artificial Intelligence, Robotics, and Automation for Space "i-SAIRAS 99", Noordwijk, The NetherlandsAn exampleof the Forwardwhere:navigationnominalplacesof interest,and ajectorystatusand pilotingnon-nominaland generatesfast executionA similarstructurearmsor erecoverypartmotion,usingand planningandnavigationto findnominalDuecontrolon MORCAnon nvironmentupdatesrocksAdestination,withfeatures,to navigation;or loose sand,of the architecturein layersandcomputation.mission.powerin the appendixrolmt,for localization,to the modulatityparallelof the sameamongis giventhe mobilereferencescan be used for the controlandlocalizesstate, and providesworld modelhazards,for exampleunexpectedstrategies.thermal,in ICA is basedthe map;internaldetectsis possibletakingS/Ssmapby the pilot; pilotingof MORCAdifferentthestrategiesand roverfeedbackthe stemAlsocontrol)functionsof otherinsideeachis supported.in differentas will be explainedelements,elementIn thismissionsuchas ontrolnext,iI2.4. FunctionalA Longplatform,RangeControlMarsmanipulator,has its own controlArchitectureRoveris a complexpointablesystemfor a Longcameras,(see FigureRangespacecraftScientificcontainingand scientificMarsRovera set of onComponents(e.g. Lander,nRover)Figure3. SpaceElementsof a LongRangeScientificAleiandro Martin-Alvarez/ 7/25MarsRovereachsuchas a mobileof theseelements
Submittedto: ArtificialIntelligence,Robotics,and Automationfor Space "i-SAIRAS99", Noordwijk,In addition, as a regular spacecraft,each control system is divided into severallocomotion/propulsion,thermal,power, and telecommuniation(see Fig. 4).)'- Accelero- ,., .v.- Voltag;s- Currents.motors-PositionFigurethe MORCAtransferArchitecture7 BodyAntenn.,-4. Spacecraftand ICA designamongfor a LongMotionrover,- ientificMarsRover.Motion c x,y navigation Value Set Parameter Value Turn Left TraverseRocky m Pilot goto x,y Pilot goto directframe)Pilot headsqueezePilot facecycle7distance (in panoramaframe) x,y (in panorama theta,x x,y (in panoramaMotionCommands n Barnsto HeadingTurn Towards h x,y Table 3. Sojournerand RockyAleiandrofrom(degreesfrom North, distance)Pilot headdirect theta,x mode7 BodyMartin-Alvarez/ 8/25forof the commandsFunctionalthe set of Sojourner n BamsTurn RightTurnwithin n Counts Maximum3 showsthePilot update x,y,z,theta (in panoramaframe, degreesNorth) n CountsSet ParameterTabledefinitionto defineControlatExposure t , Return Region from(rl,cl)to (r2,c2) with APID id Move Forwardis required'ICapture Image with CameraMaterial AdherenceMove Backwardfirst an exhaustive[20]as an example.Sojournerto Waypoint atMinutescontrol layer, called micro-RoverCommanderand to coordinateall the rover owingan ICA structure,a centralizedmissiona Mars rover scenario,has to be added to commanddatasuch-Solar PanelActuators- Receptor- Signal Power .andsubsystems,- SteeringZ-FollowingThe Netherlandsframe)Controland Rocky
Submitted to: Artificial Intelligence, Robotics, and Automation for Space "i-SAIRAS 99", Noordwijk, The NetherlandsOnceall the Commandsthe controlfunctionswere definedfor each element(see Table 4) are identifiedcontrolfollowingsystemthe erPlanningControl Wheel LevelPlanningControl ActuatorLevelPlanningControl& SensorPower ControlActuator& Sensor LevelTelecomControlActuator& Sensor LevelthereneededPlanning ControlPlanning ControlPlanning olControlfunctionsan OperationalS/Ss (e.g.Scientificplatformfor a SpaceinteractionsamongArchitectureHereLongthem2 definesMarsRoverlocomotionof commands.Architectureare located(either on-boardor on-ground)scenario.An OperationalControl ArchitectureRangemobilerefinementand & Sensor Levelfor a LongControl LevelAction LayerActuator & Sensor Level4. Controlcontrol ControlTask Layerare roverArchitecturePlanningMastControlof the ercomplexthan othersneedinga higherphilosophyare succesfulyapplied.Once & Sensor LevelTable3. DesignPlanningControlActuatorAction LayerActuator & Sensorof MORCA.Body LayerThermalControlTask Layerthen allPath principleLayerPlatformControlMicroRoverControland for its edall theseto achievethe desired controlalso defines extra operationalthat are moredesignRoverin thecontrolFunctionalfunctionsautonomyfor a spacesupport functions(seeFig. 5).2 Notation: The two parallel lines for data store, discontinousline for syncronization signal and arrow for data flow.Aleiandro Martin-Alvarez/ 9/25
Submittedto: ArtificialIntelligence,To Ground pRobotics,and Automationfor Space "i-SAIRAS99", Noordwijk,The NetherlandsFrom Groundmen SegmentFigure5. Operational3.,1. athecommunication.on queueFigureincludeand Informationand informationonesControltransferpointers,are basicallye.g. a ndIn addition,thesequeuesdesignedas singleand a executiontheOutput)lastand stInputmainlyusedIntelligenceand queuesmaintypesFirstforwhereof thavefirsteventbasedmanipulation.6 showsthis inter-communicationin a control/planningthe name of the informationto transfer(e.g. Data andassociateddiscontinousline is a signalinformationis stored in FIFOfollow a LIFO structure.queuesor stimulusexceptto representmodule.The two parallellinesCommandsParameters)and itsan eventthe non-nominalones(Input(e.g.a Command).1.4 & Output"Interactionwithother element/devicecontrollers.To / from higher level"Req CFI""Rep CFI"L1InformationExchange/Action Request"Output 1.5"CFl Status"ttor"CFI Cmd" actlV.,Input 1.5laRt .Locomotion.To / fromNominalInput 1.2Feedback"NF CFI"'---1 I"NNF CF 1" or "Fault Detected"Planning orControlOutput 1.4(e.g. ControlFunction 1 (CF1))To / from Non NominalFeedbackOutput 1.2"CFN Status"Output 1.3"CFN Cmd"Input 1.4Input 1.3From / to higher levelFigure6. IntercommunicationsAleiandroin a Control/PlanningMartin-Alvarez/ 10/25FunctionThis1.4) that
Submitted to: Artificial Intelligence, Robotics, and Automation for Space "i-SAIRAS 99", Noordwijk, The NetherlandsThecompleteScienceset of on-boardRover[ 1][20]Controlare portin the followingtable ionsthe MORCAand3.2. CapabilitiesScience pervisorSystem AdministratorControl (see Table 4)Micro-Rover5. On-boardLongphilosophy.CommandMangementControl OperationLogistic ModuleTablefor a SpacedesignRoverfunctionsfor the eArchitectureof a SpaceLongRangefA scontrolThereforeroveris mandatorytherefunctionsflexibleand Flightis a tsfor Spaceof operationto reconfigureLongRangeSciencethat the missionandsupportactuators)andcontrolthe OperationalgoalscanfailMissions.will be r,externalneededonhelp.betweentheArchitecture.In addition,this architecturereconfigurationis neededfor differentmissionphases.For example,safety is the main issue at the beginningof a space mission,placing most of control functionson thegroundsegmentwith highlyhumanintervention(humanon the controlthe mission,more challengingand highly autonomousof the control oncommunication,automaticallyautonomouswhereis also neededon-grounda ionsareloop).However,will be commandedlike softwarecompletelyand(LM);andthenLongRangefor a spaceTelemetryManagementand A),mustareControlas shownAleiandro Martin-Alvarezhavefouruseless.majorin Figure/ i 1/257.havingTherefore,the capabilityOperationat the end ofmalfunctionshas to reconfigure'its OperationalControlArchitecturein orderand find recoverystrategiesto establishcommunicationto Earth.As a conclusion,a Spacelow level commands.Forto handleroverOperationSupervisoror loss oftheto becometo receiveSupport(COS);mostrovermorehigh andFunctions:Logistic
, Submittedto: ArtificialIntelligence,Robotics,To Ground Segmentand Automationfor Space "i-SAIRAS99", Noordwijk,The NetherlandsFrom Ground SegmentTelemetryUnitCommandUnitLM Data Cmd Parameters . . .COS toCOSIaRC. InFigureThe7. MainCommandTelemetrypRC. e,downlinkof all the atedsignalsfor a sTelemetryin its inputto Earth.receivedManagementbuffer.See FigurefromgroundModuleIt also generatestohandlesa heartbeat,thetheand8.TelemetryUnitCommandCTMtoRFSH i,CommandDataBase."(,M agmm 1.--mmmFigureiCOStoiCTM l. iTelemetryDataclock\JCM to TMDataBTheUnit,;8. StructureLogisticsignalsModuleto other.of the Command(LM)containsmodules,Coma Cmd Parametersand Telemetrythe roverfor exampleManagementclocka wakeandcomputerup signal.charge of the file system makingsure that there is enoughexecutionof operationsupport and control functions.AleiandroMartin-Alvarezcomputing/ esIt sends(SA)availableis infor the
Submittedto: ArtificialIntelligence,Robotics,and Automationfor Space"i-SAIRAS99",Noordwijk,TheNetherlandsLM Data.Time & Date!ICored Cmd ParametersFig. 9. withSupervisorof the receivedthecurrentanomaliesoccurwas in "standof the Logistic(LM)operationmode;by" (e.g.duringTheof an Operational[1][23]needsof a real-timeprojects,programmers,to learn,fasterto develop,We use the An olArchitectureconceptsefficiencyControlto minimizeArchitecture,theprocesses)designalreadybut stillPlanning,andof FinitekeepingControl,care;whenwhenboththe roverpersonnel.mustbe lof controlMachine,functionsMartin-AivarezRoverthe complexityweteamsofusegraphicalthe designis quickercyclicgrouptogetherlayer/ 13/25StatedataDiagrams,Machinesandare usedprocessingincludingthat is, fastcommandDiagramsin real time,thethem.It is gofperformance.Event Handlingfunctionsfor any architectureis also used for Operationsupport functions.Aleiandroto workstructurethatArchitecture.or DispatchingStateand DataflowandScienceand interactingThereforearchitecture.prioritiesreal timewithArtificialto rol,for synchronousMachinesLongand as a consequenceof FiniteStatethe requireda StandardFinite Statedefined in the FunctionalWe esignfeedbackof the wholeDiagramsto eventsinter-communicationminimizehealththey are consistentis receivedControlcontroland less costlyfor the designreactionup signala Planetaryunderstandabilityto debug,for the use of Finiteforfor engineeringfor event-drivenreactionand Dataflowthe implementationof AI behaviors.3.3. I. ndcheckingControland wakeArchitecturemanagers,for our designmanagementfunctionof Operationalto configurationframeworkreal-timeof: resourcesthe night).usingGraphicalConcepts.of: largeis in chargefrom CM to any controllike loss of ndsworkingprocessesThus,severalwith(seeofit in parallel,(bothFiguremandatorytheis, toSystemconvenientfunctionsthatNon-Nominal,10). A subsettowereandof this
, Submittedto: ArtificialThisgroupingworkat oncein theirtakesplanningonlyRobotics,selectedbecauseFor example,its executionor anynominaladvantagefunctioncontrollayersand Automationthesethis groupingwhenevercontrolaction.the executionof this groupingto communicatevia the alreadyfor Spacefunctionscontainsis requieredin its layer cancelingA secondarycommunicationdifferentwasand ASare exclusive,the mechanismsa diagnosisat its layer,At theandis onlytime,is that all these controlwith other Finite StateoneNetherlandsis requieredeventtoprioritygenerationthe executiona non-nominalfunctionThethe desiredstrategyconsideringother controlmentionedthatNoordwijk,that establishesrecoverywithoutsameof y99".functionof nominalhasthehighestin progress.functionsMachinesshareeitherthe same interin the same orQueues.CmdlCmdNNF- Event lNNF Queue * Event NNF- Event lNF QueueUp- Report.,.Up- Report NExt- Report NDataflowforFigure10. Standarddiagramsare s onlyrequired,FiniteStateMachinein the tandaloneindependentlyto detect nominal events (e.g. a trajectoryto a Finite State Machinevia its NF queue.Aleiandrofor PlanningMartin-AlvarezFig.for continuos(includingsampled11). SometimesNominalFeedbackfinal condition/ 14/25or Controla feedback),of thisthatwholecanthat communicatesrun
) Submittedto: ArtificialIntelligence,Robotics,and Automationfor Space "i-SAIRAS99", Noordwijk,The Netherlands!iiI"CFN Cmd"FigureDataflowDiagramscomponentsevent1 1. Schematicwithto a tructureforfor a RoverNonControlNominalNNFmonitoringthatis ilitycontrolCycleSequencemaybet
of control architecture and the transition among them for a Long Range Autonomous Space Rover. We used a graphical programming for the design of the control architecture as a common and intuitive languague for a design team composed of managers, architecture designers, engineers, programmers, and maintenance personnel.
Advanced metering for SMEs The Impact of advanced metering for SMEs 0 Executive summary 02 Introduction to advanced metering 7.06 The potential benefits 06 .2 Use of advanced metering in businesses 06 .3 SupplierPrinciples of advanced metering 07 .4 Analysing advanced metering data 07 .5 Sources of energy savings 08 .6 Advanced metering technology 08 .7 Advanced metering services 09
design, testing through simulation, field implementation, and field testing of advanced wind turbine controls. This report will be Part I in a two-part series of reports that detail advanced control design, implementation, and test results. Part I (this report) will highlight the control
C5051 / C5045 / C5035 / C5030 advanced authentication advanced adobe integrations advanced storage advanced integration to enterprise systems advanced color quality advanced scanning Advanced made simple for you. . Canon’s new built-in document sharing solution simply makes sense. Multiple users can collaborate easily on a shared
Advanced Screen Systems Advanced Screen Systems Advanced Screen Systems Advanced Screen Systems Advanced Screen Systems Read completely through the installation instructions before proceeding with installation Installation requires two people Use appropriate protective equipment, including safety glasses Children should not be .
1 Gaining the knowledge of LTE-Advanced. 2 Study the physical layer of LTE-Advanced. 3 Implementation of LTE-advanced transceiver for downlink with the help of MATLAB Simulink. 4 Implementation of different MIMO configurations along with different modulation schemes and different fading channels.
during the implementation of CBEST. The data were collected through observation during the implementation of CBEST and interview with teacher and headmaster. The result of this study reveals that the implementation of CBEST has its own benefits and limitations in relation to aspect of economy, implementation and test administration and test design.
Nanotechnologies, Advanced Materials, Biotechnology and Advanced Manufacturing and Processing Part 5.ii - Page 6 of 121 Introduction In this part of the Work Programme, LEIT-NMBP stands for 'Leadership in enabling and industrial technologies - Nanotechnologies, Advanced Materials, Biotechnology and Advanced Manufacturing and Processing'.
Pearson Edexcel Level 3 Advanced Subsidiary and Advanced GCE in Statistics Statistical formulae and tables For first certification from June 2018 for: Advanced Subsidiary GCE in Statistics (8ST0) For first certification from June 2019 for: Advanced GCE in Statistics (9ST0) This copy is the property of Pearson. It is not to be removed from the
