AIAA 95-3686 ASTRO-2 Spacelab Instrument Pointing System ... › archive › nasa ›...

JlfAA

/A/NASA-CR-200148

f.

AIAA 95-3686ASTRO-2 Spacelab Instrument PointingSystem Mission PerformanceF. C. Wessling, III and Dr. S. P. SinghMcDonnell DouglasHuntsville, AL

(NASA-CR-200148) ASTRO-2 SPACELAB N96-18524INSTRUMENT POINTING SYSTEM MISSIONPERFORMANCE (McDonnell -Dougl asAerospace) 12 p Unclas

G3/19 0099869

AIAA 1995 Space Programs andTechnologies Conference

September 26-28,1995/Huntsville, ALFor permission to copy or republish, contact the American Institute of Aeronautics and Astronautics370 L'Enfant Promenade, S.W., Washington, D.C. 20024

https://ntrs.nasa.gov/search.jsp?R=19960012287 2020-07-07T08:04:11+00:00Z

ASTRO-2 SPACELABINSTRUMENT POINTING SYSTEM

MISSION PERFORMANCE

Francis C. Wessling, III

Spacelab Design EngineeringMcDonnell Douglas Aerospace,

Huntsville, Alabama 35806

Dr. S. P. Singh

Spacelab Engineering AnalysisMcDonnell Douglas Aerospace,

Huntsville, Alabama 35806

Abstract

This paper reports the performance of theInstrument Pointing System (IPS) that flew on theNational Aeronautics and Space Administration(NASA) ASTRO-2 Spacelab mission aboard theSpace Shuttle Endeavour in March 1995. The IPSprovides a stabilizing platform for the ASTRO-2instrument payload complement that consists ofthree main experiments (telescopes). Thetelescopes observe stellar targets in the universewithin the ultraviolet portion of the electromagneticspectrum that must be observed from beyond theearth's atmospheric filtering effects. The threemain experiments for observation are the HopkinsUltraviolet Telescope (HUT), the UltravioletImaging Telescope (UIT), and the WisconsinUltraviolet Photo-Polarimetery Experiment(WUPPE). The HUT uses spectroscopy to obtainthe structure and chemical makeup of ultraviolettargets. UIT is responsible for wide fieldphotographing to capture the hidden view of theultraviolet universe. The WUPPE gathers data onthe polarization of the ultraviolet electromagneticenergy coming from the astronomical targets. Thecapability of IPS enables the experiments to "see"faint celestial objects. A brief explanation of theIPS is given followed by a review of engineeringefforts to improve IPS performance over theASTRO-1 mission. The main focus ofimprovements was on enhancing the staracquisition capability through improved guide starselection, lab simulations, computer upgrades,data display systems improvements, and softwaremodifications. A star simulator was developed inthe lab to enable IPS to be simulated on theground pre-mission with flight hardware and

software in the loop. The paper concludes withresults from the ASTRO-2 mission. The numberof targets acquired and the IPS pointingaccuracy/stability is reported along withrecommendations for the future use of theInstrument Pointing System.

1ES

IPS is basically a 3-axis gyro-stabilized andcontrolled platform on which scientific instrumentsare mounted for pointing to inertial targets. ForASTRO-2, the scientific instruments wereultraviolet telescopes (HUT, WUPPE, and UIT).To meet science data gathering requirementsduring the mission, IPS inertial attitude(orientation) needed to be known and stabilizedprecisely (within arc-seconds) because gyro driftscause the IPS attitude to get corrupted. AnOptical Sensor Package (OSP), consisting ofthree star trackers, was provided to trackstationary stars and compensate for gyro drifts.

The three trackers were mounted as follows: Onetracker was designated as boresight (tracker 1),and defined IPS line of sight. Two skew trackerswere designated right (tracker 2), and left (tracker3) respectively, with lines of sight pointingnominally 12 degrees (for stellar missions) oneither side of the boresight. A payload providedAstros Star Tracker (AST) served as back-up forthe OSP boresight tracker in sensor substitutionmode. The AST was the primary star tracker forImage Motion Compensation System (IMCS)which was a part of the payload.

IPS platform is dynamically isolated from theorbrter by three torquer controlled gimbals. Thetorquers are commanded by a control system thatmaintains IPS inertial attitude while the orbftermoves during disturbances such as crew motions,and orbrter thruster firings. In addition to thegyros, an Accelerometer Package (ACP) providesfeed forward signal to the control loop. The ACPcompensates for IPS rotations caused by linearmotions (linear accelerations) of the orbiter. IPSmaneuvers are accomplished by a control systemresiding in the Data Control Unit (DCU). Controlsystem sensors are: Gyros, ACP, star trackers,and resolvers. The Control system actuators aretorquers. Figure 1 shows the IPS.

Jfl. JTA« THACKt* (ASH

IfS OPTICALPACKAGE

ASTRO-2 Payload, by Mike Lower/

Photos supplied by NASA

IPSonASTRO-1.STS-35

Generic IPS supplied by DORNIER

IPS on ASTRO-2, STS -67

Figure 1. Instrument Pointing System

American Institute of Aeronautics and Astronautics

The discussion of the IPS may be based on sevenareas. These areas are the Gimbal StructureAssembly, Payload Retention Latch Assembly,Power Electronics Assembly, Data ElectronicsAssembly, Attitude Measurement Assembly,Thermal Control Assembly, and System Software.Each area is responsible for specific functions thatallow the IPS to operate successfully.

Gimbal Structure Assembly

The Gimbal Structure Assembly (GSA) consists ofa. Static Structure, Moving Structure, Stow andUnstow Mechanisms, along with Special Devices.The Static Structure consists of a base plate,pedestal, and supporting framework. The baseplate attaches the IPS to the Spacelab palletwithin the Space Shuttle payload bay. Attached tothe base plate is the pedestal and supportstructure that provides the required rigid base forIPS. The Moving Structure consists of the DriveUnits, Yoke, Equipment Platform, PayloadAttachment Ring, Payload Attachment Flanges,and Payload Support Structure. The Yoke housesthe elevation, cross elevation, and roll drive unitsthat allow IPS to point at targets. The EquipmentPlatform is used to mount different hardwarecomponents such as the gyros. The instrumentpayload, on the Payload Support Structure,attaches to the IPS through the use of the PayloadAttachment Ring and Payload AttachmentFlanges. The Stow and Unstow Mechanismsconsist of the Gimbal Latch Mechanism andGimbal Separation Mechanism. Thesemechanisms consist of motors, actuators, andshafts that are used to latch and unlatch the IPS.The Special Devices consist of bumpers, guidehorns, plunger end-stops, Roll Drive Unit end-stopand Viewing Envelope Modifiers to keep the IPS inthe proper operational cone so ft will not impactother equipment.1 Special devices also cover theJettison Hardware that is used to eject the IPSfrom the Orbiter payload bay in case the IPS willnot stow properly.

Pavload Retention Latch Assembly

The Payload Retention Latch Assembly isresponsible for securing the payload during theascent and descent portions of the flight. This isaccomplished through the Payload AttachmentRing being separated from the EquipmentPlatform and the payload being placed into thepayload retention latches integrated onto a

Spacelab pallet. The payload must be securedduring the ascent and descent to avoid damage tothe static structure of the IPS.

Power Electronics Assembly

The Power Electronics Assembly consists of thePower Electronics Unit (PEU) and theContingency Control Panel (CCP). The PEU isresponsible for distributing the power required bythe IPS. The PEU drives the heater mats withinthe Thermal Control Assembly. The PEU alsocontains circuits for the torquers, stow and unstowhardware, and jettison hardware. The softwareobtains information about the PEU and otherdifferent systems through the use RemoteAcquisition Units (RAU). The Contingency ControlPanel is used if loss of software control occurs.The CCP can be used to stow the IPS or erectand jettison the IPS if it can not be properlystowed.

Data Electronics Assembly

The Data Electronics Assembly is composed ofthe Data Control Unit (DCU) the Data Control UnitConverter (DCUC) and the Accelerometerpackage. The DCU is a microcomputer thatcontrols the data to the IPS subsystem computers.It accepts data from the Gyro Package,Subsystem Computer, and the AccelerometerPackage to update the position of IPS. SeeFigure 2. IPS Control System Diagram. Thecontrol system is composed of two timing loops.The fast loop obtains gyro, resolver, andaccelerometer data at 100 Hz to quickly correct forposition drift while the slow loop obtains datathrough the subsystem computer from the OpticalSensor Package at 25 Hz to correct for gyro driftswithin the system.1 The DCUC provides therequired regulated power to the DCU. TheAccelerometer Package is mounted just above thepedestal and contains accelerometers in theorthogonal x, y, and z axis that provide fast loopdata to the DCU.

Attitude Measurement Assembly

The Attitude Measurement Assembly consists ofthe Optical Sensor Package and the GyroPackage. The OSP contains three Rxed HeadStar Trackers (FHST) that are used to acquirestars and report the IPS posrtion to the subsystemcomputer through the IPS RAU. There is a


IPS CONTROL SYSTEM DIAGRAM

TargetsQ

Torquer Movement

DRIVE UNITS- Elevation Drive Unit- Cross Elevation Drive Unit- Roll Drive Unit- Resolvers

Torquer Power

Power Electronics Unit(PEU)- Power- Stow / Unstow Circuits- Jettison Circuits

Gyro Package- (x, y. z, and q ) Gyroscopes- Gyroscope Electronics

IPS MOTION ]

ORBITER MOTION

Accelerometer Package(ACP)- ( x, y, z) Accelerometers- ACP Electronics

Torquer Commands

Data Control Unit(DCU)- Fast loop Software- Slow loop Software- Timing Clock

Fast Loop Data

Power Data

Fast Loop Data

Optical SensorPackage(OSP)

Timing Clock

OSPCommands

Astros StarTracker(AST)

AST CommandsSensor SubData

Target/Star Data

Remote Acquisition Unit(RAU) IPS RAU- Commands to OSP- Data from OSP- Gyro Package Interface-ACP Interface

Command and Data Bus

Computer Overide Commands

Contingency Control Panel(CCP)- Emergency Stow Switches-Jettison Switches

Input / Output Units(IOU)

SubsystemData and Commands

Note: Refer to Spacelab drawingsfor connection details.

V V

Sub System Computer(SSG)- Subsystem Computer

Application Software

ExperimentData and Commands

Experiment Computer(EC)- Experiment ComputerApplication Software

Remote Acquisition Unit(RAU) Experiment RAU- Commands to AST- Data from AST

Remote Acquisition Unit(RAU) Subsystem RAU- MFC Interface-PEU Interface- Other Interfaces Not Shown

IPS Manual Control

Manual Pointing Controller(MPC)- Astronaut Manual Interface

Payload and General SupportComputer (PGSC)- Astronaut' ap Top Interface

Mass Memory Unit(MMU)- Target Data- Mission Dependent Parameters

Figure 2. IPS Control System Diagram|


boresight FHST that is positioned along the line ofsight of the experiments along with a right and leftFHST that are positioned at a 12° separation tothe left and right of the boresight FHST. Thesubsystem computer uses the OSP information toupdate the gyro drift factor. The Gyro Packagecontains gyroscopes that are positioned in theorthogonal x, y, z, and skew "q" axes andelectronics required for communication to theDCU. The fourth gyroscope (q) acts as a backupin case of an x, y, or z gyroscope failure.

Thermal Control Assembly

The Thermal Control Assembly consists ofpassive and active thermal devices. The thermalprotection is required due to the extremetemperature variations of space. The PassiveThermal Control System consists of multilayerinsulation blankets, second surface mirrors,emissive coatings, and thermal standoffs. Theactive Thermal Control System consists of heatermats and the Spacelab Freon Loop. Temperaturemeasurements are made through the use ofthermistors that have conditioning circuits withinthe PEL) or DCU. When a system needs to beheated the heater mats are switched on to raisethe temperature of the hardware. The SpacelabFreon Loop is used to cool equipment.

IPS System Software

The IPS System Software consists of Status andMonitoring Software, IPS Applications Software,and Attitude Control Software. The Status andMonitoring Software monitors inputs from thehardware and alerts the astronauts if a parameteris out of nominal limits. The Applications Softwareis responsible for thermal control, memory access,DCU and SSC communication, and gimbal holdthat keeps IPS in position using the gimbal anglesreceived from the resolvers. The Attitude ControlSoftware is the software that is responsible for thepointing, tracking and slewing of the IPS.' TheAttitude Control Software is divided into the fastloop software and slow loop software as discussedearlier. IPS motion is controlled by the softwareusing drive units in the elevation, cross elevation,and roll axis. The software obtains data from thegyros and accelerometers to issue torquecommands to the drive units to keep the IPSpointing on target. The optical sensor package isused to obtain position information of stars for the

software to update the drift factors within thegyros. A mode of operation called sensorsubstitution is also available. Sensor substitutionuses the experiment star tracker called the AstrosStar Tracker (AST) to replace the data comingfrom the boresight FHST.

IPS Target Acquisition and Fine Pointing Mode-;

Target acquisition and fine pointing by IPS isaccomplished using one of three modes:Operational Identification (IDOP -- automaticmode), Manual Target Acquisition followed byLock On (any) Target (MTA/LOT), and SensorSubstitution. During IDOP, IPS softwaredetermines if one of pre-selected guide stars hasbeen acquired. This mode of acquisition, whensuccessful, is fast and accurate. During MTA/LOT,the crew manually acquires a target using HUTvideo, maneuvers IPS using Manual PointingController (MPC), to place the target at the centerof instrument field of view, and commands IPS tolock on any acceptable target in the field of view.

The successful operation of the IPS requires thefunctionality of the Gimbal Structure Assembly,Payload Retention Latch Assembly, PowerElectronics Assembly, Data Electronics Assembly,Attitude Measurement Assembly, Thermal ControlAssembly, and System Software. Each portion ofthe IPS has its own unique responsibilities thatallow for the IPS to maintain a stable platform forthe experiments.

IPS Improvements

Following ASTRO-1, McDonnell Douglas madeseveral improvements to IPS. They upgraded theSpacelab subsystem computer to an IBM AP101,developed a Star Simulator in the SoftwareDevelopment Facility (SDF) that enabled pre-mission IPS simulations with the flight startrackers in the loop, added the capability ofManual Target Acquisition (MTA) followed by LockOn (any) Target (LOT), improved SensorSubstitution mode, improved Objective Loadsgeneration performance to provide superior guidestar targets, enhanced the Attitude DeterminationRlter (ADF) gain sets, optically calibrated the startrackers using ASTRO-1 data, and supported startracker alignment measurements and payloadinstrument alignments prior to the mission. Inaddition, a generic Spacelab change was made


with the replacement of the Data Display System(DOS) with the Payload General and SupportComputers (PGSC) -- laptop computers.

SDF Simulations

The Software Development Facility (SDF)provides a test bed for IPS simulations. Amathematical model of IPS is used with flight typehardware in the loop to verify the flight hardwareand software functions prior to the mission. Thestar simulator enabled the flight Optical Sensorpackage consisting of three star trackers, and theexperiment optical sensor, Astros Star Tracker(AST), to be in the loop for IPS simulations.

Star Simulator

The star simulator proved to be a valuable toolthat contributed to the success of the ASTRO-2mission and is discussed further.

The Star Simulator was designed to provide alaboratory testbed for the check-out of a newlydeveloped communication and commanding modeof the Optical Sensor Package (OSP) for use onthe ASTRO-2 Spacelab mission. The ASTRO-1experience revealed the need for a manualcommanding technique for the searching andtracking of stars. The star fields were to berealistic, generated from a star catalog, ormanually created to perform specific functions inthe testing and checkout of the new flightsoftware.

The Star Simulator provides three independentstar fields measuring approximately 2.5 x 2.5degrees. The star fields are displayed on threeextremely high resolution monitors. The spectraloutput of the monitors was designed to coincidewith the spectral response of the Fixed Head StarTrackers in the OSP. The relative visualmagnitude (Mv) range required by the Rxed HeadStar Trackers is 2 to 8 Mv, which is accomplishedby a mutual calibration and adjusting softwarebrightness settings. The Star Simulator is drivenin a closed loop mode by a dynamic simulation ofthe instrument Pointing System and the star fieldgeneration system based on a 486DX 66 MHzpersonal computer.

The Star Simulator Setup consists of threemonitors for star generation, a monitor controller

to interface the monitors to a 486DX 66 MHzpersonal computer (PC), collimating optics tomake the pixels (stars) appear at infinity, anoptical bench for vibration isolation, and a 100Kclean room to keep the area clean and dark.Figure 3 shows the layout of the Star SimulatorTest Setup.

Each monitor provides an extremely highresolution of 4096 x 4096 pixels. A single pixelproduces a simulated star with a radial size lessthan 100 arc seconds. To maintain a near steadylight source the pixel update rate is 1000 Hz where50 stars (pixels) maximum are output on eachmonitor at a given time. These high speeds areobtained by using a vector monitor where theelectron gun jumps from point to point as opposedto a raster scan monitor. Each addressable pointhas a wavelength output of approximately 550 nmby using a P44 phosphor. This wavelength waschosen to match the center of the spectralresponse of the FHST. The pixel output intensityis controllable and mutually calibrated with theFHST to obtain a relative output over the range of2 to 8 visual magnitudes. The position controlallows a pixel movement of less than 5 arcseconds between adjacent points and the overallposition tolerance is less than 50 arc seconds.Each monitor combined with a collimating lensprovides a 2.55 x 2.55 degree field of view input toa FHST.

The monitor controller directly interfaces to themonitors. For each monitor the monitor controllerreceives horizontal position, vertical position,brightness, and monitor blank/unblank commandsfrom the 486DX 66 MHz PC. The monitorcontroller converts and transports these signals toeach monitor through coaxial cables. Changes inthe horizontal and vertical position data move apixel. Changes in the brightness data change theapparent visual magnitude of the star. Theblank/unblank command enables and disablesoutput on the monitor screen.

The 486DX 66 MHz personal computer containsthe Star Simulator software. The software allowsfor calibration and simulation control. The opticalsystem can be aligned using the calibrationsoftware. Optical field of view alignment consistsof first manual and then fine adjusting of the lineof-sight, followed by fine adjustment of therotation, and size parameters. Calibration alsoconsists of changing the intensity of pixels to


STAR SIMULATOR IN SDFCLOSED LOOP CONFIGURATION

AIR FILTERSMONITORCONTROLLER

IPS ATTITUDEINFORMATION

486DX66MHzPC <-

STAR •CATALOG '

SDF IPS EXTENSIONHOST 3260MULTIPROCESSORSYSTEM

IPS MODEL

OSP MODEL

NOTE: NOT ALLSDF INTERFACESSHOWN

IPS INTERFACEDEVICE (IPSID)

Figure 3. Star Simulator Test Setup


match the output expected from the FHST overthe intensity range of 2 to 8 visual magnitudes.After the system is calibrated a simulation canbegin. The simulation software uses a subset ofthe star catalog (star database) generated off line.This subset of stars defines the available starsduring the simulation and is called the Sky Map.The stars displayed on the monitor are a subset ofthe available stars in the Sky Map and is called theViewport. During simulation the position datachanges causing the output (Viewport) on themonitor to change. The software either receivesposition information by an external source throughan RS-232 link or the simulation operates in anopen loop fashion and produces a fixed position.The pixel movements make the FHST respond asif it were being moved and the stars (pixels) werestationary as on orbit. The output of the monitorsare sensed by the FHSTs through the collimatingoptics.

The collimating optics make the pixels on themonitor appear as if they were at infinity. This isrequired to simulate the stars at a far off distanceas on orbit. The collimation is achieved by placinga monitor in the focal plane of a lens. Each FHSTrequires its own collimating lens. The lens used inthe system is a 150 mm diameter achromatic lenswith a 3000 mm focal length. The system wasdefocused using a dim pixel (8th magnitude). Thedistance between the FHST and the monitor wasadjusted until the dim pixel was stable toapproximately 1 arc second. The final distancebetween the monitor and FHST was measured atapproximately 3060 mm.

To obtain a near 1 arc second stability, vibrationisolation had to be provided. The opticalvibration/isolation bench isolates the system fromexternal disturbances to allow the simulated starsto remain stable. The bench uses an 8 leg systemwhere each leg has a piston that is driven withcompressed air to support the table. Thecompressed air in the legs causes the opticalbench to 'float'. The optical bench is contained ina clean/dark room.

The monitors, monitor controller, 486DX 66 MHzPC, Star Simulator software, and collimatingoptics are the core of the Star Simulator. Theseitems form the stand alone portion of the entiresystem. The core components of the StarSimulator could be transported to another facilityand integrated with user supplied trackers, optical

bench, and clean/dark room. This would provide ahardware in the loop test bed for other trackers. Auser supplied computer could transmit the attitudeinformation data stream to the Star Simulator andreceive the data output from the user suppliedtrackers to create a closed loop test bed.

The two primary design concerns were the correctsimulation of the star movement and itsassociated brightness. The Star Field Stimulatorhad to produce simulated stars that were withinthe position tolerance and brightness range of theFHST. In brief, the simulated stars had to have amovement capability down below 5 arc secondsand position accuracy on the order of 50 arcseconds. The simulated stars had to have avisual magnitude range spanning three visualmagnitudes (Mv) in the range from 2 to 8 Mv. TheStar Simulator exceeded the requirements formovement, position, and brightness control. TheStar Simulator produced simulated stars thatallowed the FHST to behave as if it were in orbitobserving actual stars. The Star Simulator hasproven to be a valuable tool in determining theresponse of the Fixed Head Star Trackers to bothnormal and extreme operational scenarios. Testsallowed for the fine tuning of the applicationsoftware. This tool proved its usefulness over thatof a mathematical model of any fidelity because ofthe invaluable hardware-in-the-loop testingcapability. This was particularly evident during theASTRO-2 mission when it showed a softwarepatch for moon operations would not have workedif it were uplinked to the crew.

The use of the IPS Star Simulator during themission allowed the engineers in the lab on earthto verify procedures before being uplinked to thecrew of Endeavor. Specifically, during the earthmoon observations, the brightness of the moonwould damage the boresight Fixed Head StarTracker. As a result, alternatives were eliminatedby testing in the lab using the spare flight OSPintegrated with the IPS Star Simulator in the SDF.Tests showed that turning down the high voltage,instead of closing the shutter by tracker reset,resulted in a reduction in the count rate from only4500 to 2900 counts. Thus, the tracker would stillbe damaged by the moon. As a result of testing inthe lab, the decision was made to turn off theboresight tracker during the remaining moonobservations.

8American Institute of Aeronautics and Astronautics

Mission Performance Results

Targets

1.

2.

3.

The ASTRO-2 scientists made over 300observations of 236 targets in the ultra-violet,range of the electromagnetic spectrum. Thisincluded distant quasars, stars, super novas, andgalaxies. The planets Jupiter, Mars, and Venuswere observed. In addition, Jupiter's moon lo, andour own Earth's moon were viewed.

Lunar Observations

IPS was designed to point to inertially fixedtargets, but during ASTRO-2, the capabilities ofIPS were stretched. It was used to track a movingtarget.

During the ASTRO-2 mission, four lunarobservations were made. Observing the moon isdifferent from observing stellar targets becausethe moon's inertial attitude changes continuouslydue to its relative nearness (parallax) to earth,while stellar targets are inertially fixed. Also, themoon is too big for an IPS tracker to acquire andhold optically. Furthermore, the moon is too bn'ghtfor the IPS trackers, and its light could damage theimage dissector tube (IDT) of the tracker.

These considerations resulted in the following:

IPS attitude would need to change to stay pointedat the center of the moon. IPS has been designedto fine-point to inertially fixed targets. If IPSattitude was held fixed, the moon would appear tostreak across the instrument fields of view. Tokeep the observing telescopes pointed to thecenter of the moon, IPS would need to track thecenter of the moon.

The moon could not be used as a guide starbecause it projects too big of an image on thetracker's IDT. To point to the center of the moonwith the desired pointing accuracy, a two-skew-tracker IDOP would have to be performed.

The boresight tracker's shutter would need to havebeen closed prior to the slew to the moon attitude,to protect the boresight tracker from possibledamage to its image dissector tube.

The method chosen for moving IPS may bethought of as a two-step process. The first step

was to command IPS to point to an inertially fixedattitude. The commanded inertially fixed attitudewas along the vector joining the center of themoon to the center of the earth at the start ofmoon tracking. The second step was to add atime varying experiment bias command to moveIPS to point to and stay pointed at the center ofthe apparently moving moon. The initial value ofthe time-varying bias pointed the IPS line-of-sightto the center of the moon and the subsequentvalues maintained the track.

This method was chosen because IPS softwaredoes not have a provision to continuously changethe commanded attitude parameter. Thecommanded attitude and changing experimentbias together brought about the desired change.

The high voltage for the IPS boresight trackermust be turned off to avoid damage to the trackerdue to moonlight. For the first observation, thiswas accomplished by issuing a reset of the trackerthat puts the tracker in an unloaded state. Forsubsequent observations, the procedure waschanged to power off the boresight tracker.

Moon attitude was achieved as follows: A two-skew-tracker IDOP was performed at the moonattitude. When the fitter settled, optical hold wasturned off (IPS was put under gyro control), target-track bias was enabled, and the moon wastracked.

IPS Pointing Performance - Accuracy and Stability

IPS pointing performance was evaluated for a fewtypical observations. A visual output of the IPS lineof sight is shown in Figure 4.

-49J

Ifl-

9CT $S TZ SOT HAWAII IK W)T VEVPIT- 0US0706W

:.::::-j;|li!J!!|n|;:;:i:.

r̂ .m -,«...«.,«,. "M"ota> **""

J "if

Figure 4. Boresight YZ Spot Diagram


Each spot on Figure 4 represents a data sampletaken from the y and z coordinate position datareported from the boresight Fixed Head StarTracker. The horizontal and vertical scale is 1 arcsecond per division. Pointing accuracy measuredas the mean pointing error from commandedattitude was 0.85 arc seconds.

The stability of the IPS is evaluated using theOptical Sensor Package (OSP) coordinatemeasurements during periods of optical hold.Table 1 gives the standard deviations of OSP line-of-sight (LOS) measurements over several 200second quiescent periods. The quiescent periodswere selected to occur between orbiter thrusterfirings. A sub-arc second quiescent LOS pointingperformance was typically maintained, and often

exceeded the 0.75 arc second quiescent pointinggoal.

Sensor substitution was only employed on a fewoccasions, but the LOS pointing performanceappears to show a slight improvement asexpected due to the lower noise characteristics ofthe AST compared to the FHST. Table 2 givesthe standard deviations for the sensor substitutionutilizing the Astros Star Tracker (AST)measurements in the Attitude Determination Filter(ADF). A definitive statement of the improvedperformance would have required more sensorsubstitution operations.

ASTRO-2 IPS QUIESCENT POINTING PERFORMANCEStandard Deviation (arc seconds)

(Between thruster firings)

GMT(RelativeSeconds)0.66:14:10(350-500)066:14:40

(1400-1500)066:19:10(1 00-300)068:04:40

(1200-1400)069:06:40(1 00-200)

Boresighttrackery-axis0.71

0.83

0.85

Boresighttrackerz-axis0.67

0.77

0.68

Righttrackery-axis0.71

0.64

0.72

0.61

0.83

Righttrackerz-axis0.68

0.66

0.69

0.63

0.72

Lefttrackery-axis0.80

0.65

'0.76

0.73

0.88

Lefttrackerz-axis0.83

0.65

0.71

0.67

0.70

Table 1: ASTRO-2 IPS Quiescent Pointing Performance

ASTRO-2 SENSOR SUB PERFORMANCEStandard Deviation (arc seconds)

(Between thruster firings)

GMT (relative seconds)070:18(330-430)075:20(8,150-8,250)073:11 (14,750-14,850)072:11 (4,350-4,550)

Y0.650.890.420.42

Z0.500.910.430.48

Table 2: IPS Sensor Substitution Quiescent Pointing Performance


Summary

The Instrument Pointing System (IPS) performedsuperbly for the 14 plus days of IPS operationsduring the ASTRO-2 mission. IPS performedbetter than expected. The IPS control systemfunctioned perfectly, and ADF performance wassuperior. The pre-flight predictions of alignmentMDPs agreed closely with calibrations performedduring the mission. The count rate predictions ofguide stars for the IPS star trackers consistentlymatched the measurements observed during themission. MTA/LOT was totally successful. IPSstability under optical hold was superior comparedto ASTRO-1, as shown by a marked contrast withthe one observation in which the crew performedmanual hold. The pointing stability and accuracywere both less than 1 arc seconds.

NASA Headquarters, MSFC, Johnson SpaceCenter, Principal Investigator (PI) teams, thepress, general public, and MDA managementexpressed delight with the better than expectedsuccess of the mission. The Pis in particular wereecstatic. There was overwhelming response tointernet coverage of the mission. There were over2.6 million accesses of the home pages fromabout 200,000 individuals from 59 countries.

IPS is a versatile platform for fine-pointing atheavenly objects, and should be considered forfuture use.

ACKNOWLEDGMENTS

Our thanks to the National Aeronautics and SpaceAdministration (NASA) Headquarters and theNASA Marshall Space Right Center (MSFC) fortheir efforts that led to the successful ASTRO-2flight. The authors acknowledge the contributionsof TRW for software development, SODERN forthe technical support, and Orwin Associates, Inc.for the development of the monitors and monitorcontroller. Thanks also go to McDonnell Douglasfor giving the authors the freedom to get the jobdone.

References

1. JSC, "Instrument Pointing System SpacelabSystems Training Manual". National Aeronauticsand Space Administration Lyndon B. JohnsonSpace Center, S IPS 2102.

2. Wessling, III Francis C. and Vander Does,Mark, A., The Star Reid Simulator For TheSoacelab Instrument Pointing System Fixed HP^HStar Trackers. SPIE-The International Society forOptical Engineering, Proceeding Volume 2221,Acquisition, Tracking, And Pointing, April 1994.

3. MDA, "Design Specification for the InstrumentPointing System (Igffl Star Simulator". McDonnellDouglas Aerospace, Drawing 9007627.

4. MDA, "IPS Star Simulator". McDonnell DouglasAerospace, Drawing 9007626.

5. Orwin Associates, Inc., "Operations andMaintenance Manual for the Orwin AssociatesInc. Model 1534 Hioh Speed Graphics Disnlav"Orwin Associates, Inc.

6. MDA, "IPS Additional Tasks Final Report.Part I". McDonnell Douglas Aerospace, SLMDH-0306.


AIAA 95-3686 ASTRO-2 Spacelab Instrument Pointing System ... › archive › nasa ›...

Documents

Transcript of AIAA 95-3686 ASTRO-2 Spacelab Instrument Pointing System ... › archive › nasa ›...