gp-b attitude and translation...
TRANSCRIPT
-
11
GP-B Attitude and Translation Control
John MesterStanford University
-
22
Space - reduced support force, "drag-free"
- roll about line of sight to star
Gyro - sphericity and mass unbalance
requirements: 10nm achieved
The GP-B Challenge
– Gyroscope (G) 107 times better than best 'modeled' inertial navigation gyros– Telescope (T) 103 times better than best prior star trackers– G – T
-
33
ATC Requirements
Pointing Requirement 110marcsec/√ Hz (Guide Star Valid) pitch and yaw
Roll Requirement 40 arcsec at roll rate
Acceleration Requirement Active Control of all 6 dof of Spacecraft
-
44
What is the ATC system?
First space vehicle to actively control all six degrees of freedom 3 Orientation 3 Translation
16* Cold gas proportional micro thrusters provide forces and torques Fueled by helium boil-off gas from cryogenic system * Two thrusters failed before science phase
Common-mode flow rate control maintains helium bath temperature
Pointing system controls the guide star tracking telescope and maintains roll phase Translation control system uses acceleration measurements from one of the science
gyroscope's suspension system to null out environmental forces Vehicle flies in a near-perfect gravitational orbit
-
55
Block Diagram
Rate Sensor
Orientation Sensors
Command Generator
Attitude Control
Translation
Control
Dewar Temp
Control
Thruster
Distribution
Magnetic
Torque
Rods
Thrusters
Vehicle
Dyanmics
Science Gyro
Suspension
System
Dewar Press
& Temp
Sensors
Magnetic Torquer
Distribution
+
-
++
Dewar
Thermodynamics
Attitude and
Translation
Control
Software
Sensors Actuators
GP-B
Spacecraft
+
-
PrimaryTelescope
Star Trackers
BackupMagnetometersCoarse Sun Sensors
Control Gyroscopes
-
66
ATC System Hardware
Sensors
• Primary Sensors• Attitude
• Telescope• Star Tracker• Rate Gyroscopes
• Translation - One of the fourscience gyro suspensionsystems
• Backup Attitude Sensors• Magnetometers• Coarse Sun Sensors
• Backup Translation Sensors -Three more science gyroscopes
Actuators
• Primary Attitude and TranslationControl Actuators - 16* ProportionalMicro Thrusters
• Backup Attitude Actuators - Magnetictorque rods
-
77
The Overall Space Vehicle
♠ 16 Helium gas thrusters, 0-10 mNea, for fine 6 DOF control.
♠ Roll star sensors for fine pointing.
♠ Magnetometers for coarseattitude determination.
♠ Tertiary sun sensors for verycoarse attitude determination.
♠ Magnetic torque rods for coarseorientation control.
♠ Mass trim to tune moments ofinertia.
♠ Dual transponders for TDRSSand ground stationcommunications.
♠ Stanford-modified GPS receiverfor precise positioning. Laserranging corner cube cross-checksGPS.
♠ Redundant spacecraftprocessors, transponders.
-
88
Control Gyroscopes
• Two gyro assemblies -one used, one backup
• Gyro mounting platformthermal distortion had abig impact on the controlsystem performance
ControlGyroscopes
-
99
Star Trackers
Star Trackers
• Two star trackers - oneprimary, one backup
• 8 deg field of view• Mounted at angles 10degrees apart - differentgroups of stars visible
-
1010
Magnetometers
• Measure vehicleorientation by comparingmagnetic field readingwith expected field
• Expected field dependson orbital position whichis propagated onboardand updated from ground
Magnetometers
-
1111
Coarse Sun Sensors
• Backup coarseorientation sensors
• Rough estimate ofvehicle orientation
• Only used if star trackersand magnetometers fail
Fwd SunSensors
Aft SunSensor
-
1212
Thrusters
• Vehicle has anominal mass flowrate of ~8 mg/s
• Maximum thrusterforce: 8mN
Thrusters
-
1313
Magnetic Torque Rods
• 3 Orthogonal torque rods• Two axis control dependingon magnetic field geometry• Orbit position propagatedonboard and updated withground processed GPS data• Thruster failure backup• Used during science toincrease fuel margin• Still used today to controlsatellite to ~5 deg
-
1414
Attitude Control
• Science Mode– Pitch/Yaw Pointing
• Telescope• Control gyroscopes
– Roll Control• Star tracker• Control gyroscopes
• Coarse Control– Pitch/Yaw Pointing
• Star tracker• Control gyroscopes• (Magnetometers)• (Coarse sun sensors)
– Roll Control• Star Tracker• Control Gyroscopes• (Magnetometers)• (Coarse sun sensors)
-
1515
Single Axis Pointing Error
0 10 20 30 40 50 60 70 80 90 100-15,000
10,000
5,000
0
5,000
10,000
15,000
North-South Pointing vs Time (11/10/04)
Time Since Guide Star Invalid Start (min)
Poin
ting
Angl
e (m
arcs
ec) Vehicle in Solar Eclipse
2) Guide Star Capture
1) Guide Star Invalid 3) Guide Star Valid
The attitude control goes through three different phases each orbit:1) Guide Star Invalid – Guide star is blocked by Earth2) Guide Star Capture – Guide star is re-centered in the telescope field ofview3) Guide Star Valid – Nominal telescope pointing
-
1616
Target Plot
-4000 -2000 0 2000 4000
-3000
-2000
-1000
0
1000
2000
3000
Spacecraft Pointing (3/25/05)
East-West Pointing (marc-sec)
Nor
th-S
outh
Poi
ntin
g (m
arcs
ec)
3600 marcsec
Nominal guide star capture times range from 30 to 90 seconds The RMS vehicle pointing is controlled to less than 320 marcsec (1.55microrad)
RMS Pointing = 260 marcsec
-
1717
Proton Monitor Data
ProtonMonitor Hits
-
1818
Telescope Detector Data
• Particle hits also affected the telescope detectors• A plot of the number of hits in the telescope clearly shows the South AtlanticAnomaly region
TelescopeDetector
Proton Hits
-
1919
SAA Data Plot
GP-B flies through the SAA at least three out of every fifteen orbits losing,at times, up to 80% of the telescope data to proton corruption
0 10 20 30 40 50 60-1000
-500
0
500
1000Vehicle X-Axis Attitude Error (5/9/05)
Time Since Guide Star Valid Start (min)
Ang
le (m
arc-
sec)
0 10 20 30 40 50 600
20
40
60
80
100Percentage of Telescope Data Corrupted by Proton Activity
Time Since Guide Star Valid Start (min)
Per
cent
-
2020
Roll Rate Plot
2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500
0
50
100
150
200
250
300
350
400Vehicle X-Axis Pointing
Orbit Number
Ampl
itude
(mar
csec
)
Total RMS PointingRoll Rate Component
• Mission-longpointingperformance
• Shaded Regions- Roll Rate NotchFilter Turned ON
-
2121
gyro output Scale factorsmatched foraccuratesubtraction
telescope output
Dither: Slow 30 marc-s calibration oscillations injected into pointing system
{Vehicle pitch/yaw dither
Once telescope/gyroscope scale factor is known, vehicle motion can be subtracted outof the gyroscope signal
Pitch/Yaw Attitude Dither
-
2222
Roll Phase Control
There are two star trackers onboard the GP-B satellite.• Field-of-view: 8ox8o• Boresight axes are 50o and 60o away from the satellite roll axis,
out of phase by 180oTwo Attitude Reference Platforms (ARPs) are mounted on the
graphite ring around the dewar of the satellite.One control gyroscope package and one star tracker are mounted on
each ARPClose-loop control of the roll phase is implemented with 16
proportional cold gas thrusters by the Attitude and TranslationControl (ATC) system.
-
2323
Blue = Star Tracker A Field of View
Red = Star Tracker B Field of View
Star Tracked Fields of View
-
2424
Roll Angle Plot
0 10 20 30 40 50 60-80
-60
-40
-20
0
20
40
60
80Roll Phase Error (3/25/05)
Time Since Guide Star Valid Start (min)
Angl
e (a
rc-s
ec)
40 arcsec
• The controlgyroscopes andstar trackers areused to control thespacecraft roll angleto an error less than40 arcsec at aconstant roll periodof 77.5 sec
-
2525
Drag-free Satellite Technology
• 1959 – George Pugh envisioned a “tender satellite” for first proposedtest of General Relativity.
• 1964 – Ben Lange (Stanford) provided first detailed study of issuessurrounding drag-free satellites.
• 1972 – Flight of a Disturbance Compensation System (DISCOS) onTransit I (experimental US Navy navigation satellite)
• 2004 – Flight of Gravity Probe BTransitpre-launch
Transit deployed on orbit
-
2626
Drag Free Concept
Control Spacecraft to follow an inertial sensorReduce disturbances in measurement band
Aerodynamic dragMagnetic torquesGravity Gradient torquesRadiation Pressure
Spacecraft Follows a purely Gravitational Orbit
-
2727
Drag-Free Satellites have flown successfully
TRIAD I : Johns Hopkins Applied Physics Laboratory Navy Transit Navigation System
Launched September 2, 1972 Polar Orbit at 750 km Mission Lifetime over one yearDISCOS - Disturbance Compensation System - Stanford 3 axis translation control
And Now Also GP-B 3 axis translation control3 axis attitude control
Drag Free History
-
2828
• Electrostatic Sensing of Proof Mass
• Pressurized gas “On-Off” Thrusters
• 3 Axis Translation Control
• Acceleration levels were below 5 x 10–12 g
averaged over 3 days
– limited by tracking data and earth
gravity model
DISCOS Performance
-
2929
Propellant Source: Helium Boil-off
Isola
tio
n V
alv
es (
TIV
) -
16
Th
rust
ers
-1
6
Pyro
vent
valve
Su
perf
luid
Heliu
m
240
0L
Heat
Leak
125mW
Poro
us P
lug –
Su
perf
luid
ph
ase s
epa
rato
rW
i ck
• Superfluid phase separator“porous plug” permitsgaseous Helium to leavemain tank while keepingsuperfluid He inside.
• Mass Flow: 4 - 16 mg·s−1 overplug temperatures of 1.6K to2.0K without choke orbreakthrough.
• Provides supply pressure of660 to 2300 Pa (5 to 17.5 torr)to thrusters.
• Each thruster is provided adedicated isolation valve(TIV) to disable unit if afailure occurs.
2.8K
1.8 K
300Kambient
-
3030
Flight Proportional Thruster Design
From Dewar and manifoldSupply: 5 to 17.5 torr
Re ~10 – Laminar flow!
Thrust: 0 – 10 mNISP: 130 secMdot: 6-7 mg·s−1
Scale factor variation: 6%
Bias variation: 0.2 mN
Noise: 25 µN·Hz−1/2
Operates under choked flowconditions
Pressure FB loop makesthrust independent of unittemperature
3.5mm dia
Redundant voice coils
-
3131
Thruster Arrangement on Vehicle
• Thrusters arearranged in 4 clustersof 4 units each.
• Each cluster hasthrust authority along3 axis.
• Arrangement chosento be robust tothruster failures.
• Provides 3DOF ofattitude and 3DOF oftranslation control forthe space vehicle.
+X
+Y
+Z
7
8
14
1
11
12
2
16
3
9
1310
6
5
15
4
1.19 m
1.19 m
1.19 m
1.19 m
2.51 m
1.9 m
Space Vehicle
Body Axes
Roll Axis
Thruster
Cluster
Guide
Star
SV
Forward
SV Aft
G1
G4
23cm
8.8
cm
-
3232
“Prime” Drag-free Operation
• In prime drag-free, the ATC system flies the SV around the position of theproof mass– Suspension control is turned “off” (up to a given safety radius)– Advantage: Suspension forces/torques minimized (somewhat)– Disadvantages: 1) unsuspended spinning “bomb”, 2) cannot correct for
accelerometer biases
( )r R!��
R� r
�SV Translation
PID Controller
Resonance
Modes
+ +
21Ms
+
Position
Bridge
Vehicle dynamics
Gyro Suspension
Controller 21
ms
Gyroscope dynamics
+
+
Gravity Gradient feed forward
GSS Ctrl
Effort
K1K2
Gyro
Position
SV
Position
+
-
-
Gyro Suspension
Translation Control
Space Vehicle Dynamics
Ext
Distur -
bances
+
Ext
Disturbances
SV Ctrl
Effort
U
R�
r�
( )r R!��
-
3333
“Backup” Drag-free Operation
• In suspended (backup) drag-free, the ATC system nulls the gyro suspensioncontrol effort.– Advantage: Gyro “always” suspended; 2) Accelerometer biases can be
removed.– Disadvantage: Suspension forces and torques reduced to preload
minimums.
( )r R!��
R� r
�
SV Translation
PID Controller
Resonance
Modes
+ +
21
Ms+
Position
Bridge
Vehicle dynamics
Gyro
Suspension
Controller2
1ms
Gyroscope dynamics
+
+
Gravity Gradient feed
forward
GSS Ctrl
Effort
K 1K2
Gyro
Position
SV
Position
+
-
-
Gyro Suspension
Translation Control
Space Vehicle Dynamics
Ext
Distur -
bances
+
Ext
Disturbances
SV Ctrl
Effort
R�
r�
u�
U�
( )r R!��
Accelerometer
bias
compensation
-
3434
Example Drag-free Transitions
Transition sequence:
1. Backup drag-free
2. Prime drag-free
3. Non drag-free.
1. Prime drag-free modenot used on orbit due tounacceptableaccelerometer biases.(20 nN on proof mass)
• Note: data shown isearly on-orbit testsequence; does notrepresent finalperformance.
0 20 40 60 80 100 120
-10
0
10
Science Gyro PositionPo
sitio
n (n
m)
0 20 40 60 80 100 120-100
-50
0
50
100Science Gyro Control Effort
Forc
e (n
N)
0 20 40 60 80 100 120-5
0
5Space Vehicle Thrust
Minutes Since 08/22/2005 18:05
Forc
e (m
N)
Vehicle XVehicle YVehicle Z
Vehicle XVehicle YVehicle Z
Vehicle XVehicle YVehicle Z
Drag free off
Drag free on
-
3535
Overall Drag-free Performance
• Performance at4x10-12 g levelbetween 0.01 mHzand 10 mHz in inertialspace.
• Suppression ofgravity gradientacceleration by afactor of ~10,000.
• Meets needs ofexperiment tominimize supportforces on gyroscope.
10-4
10-3
10-2
10-1
100
10-11
10-10
10-9
10-8
10-7
10-6
Frequency (Hz)
Control effort, m.sec
-2
Gyro 1 - Space Vehicle and Gyro Ctrl Effort - Inertial space (2005/200, 14 days)
Z Gyro CE
Z SV CE
2!OrbitGravity Gradient
Residual gyroacceleration
-
3636
Dewar
Dewar
-
3737
Dewar control plots
• The ATC systemmaintains a constantdewar temperature of~1.83 K by controllingthe flow rate of heliumboil-off from the dewar
-
3838
Dewar Heat Pulse
• The temperaturecontrol behavior isobservable during aheat pulse test (usedto estimate theamount of heliumremaining in thedewar)
-
3939
Lesson: Accelerometer biases Bends OrbitEf
fect
ive
SV Z
-axi
s th
rust
(mN
)
• Accelerometer biasdue to small patchcharges generatedrag-free force biason vehicle.
• Bias identified viaGPS and laserranging; removed viaparameter update.
• Necessitated backupdrag-free operation tocompensate forbiases; no ability tocompensate in primedrag-free.
-
4040
Lesson: Dewar slosh mode coupling
• Dewar slosh mode resonance peakpumped by drag-free controller
• Managed via drag-free loop gain settingand crossover phase margin.
• Simple fluid wave model predictsperiod with high accuracy:
2slosh
roll
l rT
v ad
rT
d
!= =
=
Periods:100 to 200 sec
(without the addition of SVroll period, 77.5 sec)
v ad=
2
rolla r!=
d
r
Wave
Velocity
-
4141
Slosh Mode Evolution
Peak at 1.2x10-7 g 217 sec period
-
4242
Conclusions
• 100marcsec/√ Hz pointing Achieved– 5marcsec at roll frequency
• Roll phase controlled to