Six Decades of Guidance and Control Systems Experiences Henry Hoffman, Swales Aerospace Systems Engineering Seminar December 5th, 2006

1
Six Decades of Guidance and Control Systems
Experiences Henry Hoffman, Swales
Aerospace Systems Engineering
Seminar December 5th, 2006
0
2
Objectives
0

• Discuss key design development issues
  associated with guidance control systems
• Discuss troubleshooting techniques to resolve
  guidance control anomalies
• Discuss on-orbit satellite ACS troubleshooting
  techniques processes
• Discuss on-orbit satellite ACS recovery
  techniques processes

3
1940s
0

• WWII RADAR and fuses
• Werner von Braun
• Wind tunnels from Peenemünde

4
1950s

• Angled arrow projectile/gun- launched guided
  missile (AAP/GLGM)
• SUBROC development test
• Explorer I (1/31/58)

R
5
1950s Explorer-1 Satellite

• First US satellite Launched Jan. 31,1958 on
  modified Jupiter-C rocket by Army Ballistic
  Missile Agency
• Spin-stabilized
• Mass 14 kg.
• Perigee 347 km.
• Apogee 1,859 km
• Inclination 33.2
• Developed by Jet Propulsion Laboratory
• Carried U.S.-IGY (International Geophysical Year)
  payload of James Van Allen
• Resulted in discovery of radiation belts around
  Earth

6
1950s Explorer-1 Satellite (cont.)
Experienced On-Orbit Flat Spin Instability Issues
Flexible Whip Telemetry Antennas
R
7
1950s Explorer-1 Satellite (cont.)

• Explorer-1 stability issues
• Conservation of angular momentum applied
• Designed to be spin-stabilized about its minimum
  inertia axis
• Rotation of rigid body about either maximum or
  minimum inertia axis can be stable
• Problem Explorer-1 not really a rigid body
• Energy dissipation in flexible whip telemetry
  antennas had destabilizing effect
• Resulted in flat spin state

8
Holy Grail of Satellite Dynamics

• Conservation of momentum must hold
• HI?K(constant)
• Energy
• E½ I?2
• Emin½ (IMAX?min)?min ½ K?min
• EMAX½ (Imin?MAX)?MAX ½ K?MAX

9
1960s Dawn of Satellite Era

• Orbiting Geophysical Observatory (OGO-4)
• Launch date July 28, 1967
• Orbiting Astronomical Observatory (OAO-2)
• Launch date Dec. 7,1968

R
10
1960s OGO-4

• Mass 562 kg
• Mission Conduct diversified geophysical
  experiments
• Zero-momentum ACS
• Problem Sun-induced thermal oscillations of
  60-ft.- long experiment antenna (floppy
  STEM-type) boom
• Created attitude disturbance that ACS controller
  responded to by firing thrusters

11
1960s OAO-2

• OAO science goals dictated stringent ACS
  requirements for pointing accuracy pointing
  stability
• ACS for OAO-1 was gyroless zero-momentum design
  due to lack of confidence in gyro reliability
  (lifetime) for multi-year mission
• Problem OAO-1 suffered catastrophic star tracker
  failure (high voltage arcing) entire mission
  was lost
• ACS had total of 6 two-axis gimbaled star
  trackers
• OAO-2 was equipped with inertial reference unit
  from MIT with high-precision gyros

12
1970s
IUE (1/26/78)

• SMM (2/14/80)

RAE-B (6/10/73)
SSS (11/15/71)
R
13
1970s SSS

• Small Scientific Satellite
• a.k.a. Explorer-45 or S3
• Spin-stabilized
• Mass 52 kg
• Inclination 3.2
• Mission Study magnetosphere
• Problem Thermal energy from sun deflected
  stiff radial booms at nutation frequency
• Resulted in nutational instability
• Demonstrated that this type of nutation
  instability is not limited to floppy booms
• SSS essentially had energy source at nutation
  frequency
• SSS nutation went away when spin axis moved close
  to sun
• Maximum boom deflection, but minimum forcing
  function at nutation frequency

14
1970s SSS (cont.)

• Fundamental nature of problem concerned thermal
  time constant of boom
• Infinite time constant would mean no thermal
  bending at any frequency
• Zero time constant would mean instantaneous
  thermal bending occurs in phase with motion
  relative to the sun (does not impact nutation)
• If time constant is such that there is some mass
  motion phase shift at nutation frequency, there
  will be impact on nutation
• Not just booms - could be any moving mass on
  spacecraft
• Impact could be stabilizing or destabilizing
• Similar nutation instabilities occurred on
  Ulysses spacecraft in 1990s

15
1970s RAE-B

• Also known as Explorer 49
• Mission Radio astronomy
• Mass 328 kg
• Orbit Lunar
• Modified design of RAE-A to include bolt-on
  propulsion module needed for transfer orbit to
  Moon
• Spin stabilized during transfer orbit to Moon
• 750'-long booms attached to central body (1,500'
  tip-to-tip)

16
1970s RAE-B Schematic
LUNAR INSERTION MODULE
CENTER OF GRAVITY
VELOCITY CONTROL PROPULSION SYSTEM
R
17
1970s RAE-B (cont.)

• Potential problem New bolt- on propulsion
  module mounted well below plane of spacecraft CG
• Small principal axis misalignment would result in
  fluid transfer, causing further misalignment of
  maximum inertia axis away from geometric axis
• Would have unfavorably impacted desired spin axis
  in transfer orbit (up to 15º from geometric spin
  axis)
• Was discovered analytically by GSFC prior to
  launch

R -16
18
1970s RAE-B Propulsion Diagram
Solution Pinching off gas interconnect prevented
any significant fluid transfer Gas fuel
interconnects allow tanks to maintain equal
pressure on both tanks in spin stabilized mode
R -16
19
1970s IUE

• International Ultraviolet Explorer (IUE)
• Developed as GSFC in-house space science mission
  
• Zero-momentum ACS
• Mission Analyze ultraviolet light from stars
  blocked by Earths ozone layer
• 3 year design life requirement, 5 year goal
• July 1982 Down to 3 operating gyros from 6 (no
  spares)
• Control scheme proposed that used FSS with any 2
  remaining gyros
• Simulated, developed ready for operation by
  Spring 1983
• Aug. 17, 1985 4th gyro failed, leaving
  spacecraft in safe hold mode with only 2
  functioning gyros
• Upload of new 2-gyro control scheme was
  successful science operations resumed
• Later, single gyro also fully tested on-orbit,
  but never required
• Zero gyro case was also simulated (see SOHO)

R -12
20
1970s SMM

• Solar Maximum Mission developed using MACS
  technology
• Mass 2315 kg
• Zero-momentum ACS
• Low-Earth orbit (574 km)
• Mission Study sun during high part of solar
  cycle
• First spacecraft designed with on-orbit repair
  capability
• By November 1980, a loss of reaction wheels
  (blown fuse) in satellites MACS module cut short
  original mission
• GSFC developed magnetic control laws to maintain
  SMM in power thermally safe condition until it
  could be serviced
• See 1980s slides for servicing mission details

R -12
21
1980s
TDRS-1 (4/5/83)
Ulysses (10/6/90)
COBE (11/18/89)
SMM Servicing (4/84)
R
22
1980s TDRS-1

• Momentum-bias ACS
• Used for relaying telemetry tracking from other
  space assets to ground, while not in view of
  ground station

23
1980s TDRS-1 (cont.)

• Inertial (interim) Upper Stage (IUS) gimbaled
  nozzle failure caused stack to spin up to
  180º/sec
• About 6 hours after launch, TDRS separated from
  IUS placed in stable inertial hold mode
• Resulting orbit was highly elliptic (perigee
  several thousand miles low)
• Operations further complicated when ground crew
  commanded TDRS into earth pointing mode, leading
  to excessive thruster firings, which overheated
  thrusters
• Recovered orbit over 4 months with station
  keeping thrusters (39 ? V burns up to 180 min.)
• 1 LBF thrusters used to de-spin, perform a total
  ? V of 1,000 fps to raise perigee 8,600 miles
  for 5,000-lb. spacecraft
• Once 1 bad thing happens, take time to think
  about recovery BEFORE making another decision!

24
1980s SMM Servicing

• Spin stabilized to 0.1/sec.
• Shuttle mission 41-C (April 6, 1984)
• George Pinkie Nelson attempted to grapple SMM
• Original grapple device didnt work
• Hand attempts to de-spin caused tumble that had
  to be recovered by ground before servicing EVA
  was successful
• 10,000 LOC reloaded to get 6 LOC for Bdot
  control to de-tumble spacecraft
• First satellite repair on-orbit
• Mission completed December 1989

25
1980s COsmic Background Explorer (COBE)

• Developed by GSFC to study residual radiation of
  Big Bang
• Zero momentum ACS
• Design philosophies
• No single point failures
• Autonomous safe hold
• Fail operational
• Fail operational philosophy used to prevent sun
  exposure of main science instrument dewar
• Minutes of exposure would result in days of lost
  helium
• Suffered gyro failure in 1st week of launch
• Came up over horizon fully operational

26
1980s COsmic Background Explorer (COBE)
Dr. John Mather of GSFC
27
1980s Ulysses

• ACS Spin-stabilized 5 rpm
• Joint ESA-NASA mission in its 17th year of
  operation
• Mission Explore polar regions of sun
• Planned mission 10/1990 9/1995
• Dimensions
• Length (booms stowed) 3.2 m
• Width 3.3 m
• Height 2.1 m
• Mass
• Total spacecraft 370 kg
• Scientific payload 55 kg
• Propellant 33 kg

Only spacecraft in near solar-polar orbit
28
1980s Ulysses (cont.)

• After launch, Ulysses operated normally in its
  cruise phase for about 1 month, with proper
  operation of passive nutation damper verified by
  sensor data
• Problem Minutes after deployment of a 7.5-m
  axial boom antenna, Ulysses began unexpected
  nutation that grew to 7º peak-to-peak
• When spacecraft operated CONSCAN earth tracking
  mode (using thrusters), nutation disappeared

29
1980s Ulysses (cont.)

• Analysis predicted thermal deformation of boom at
  nutation frequency would cause nutation based
  upon Earth-spacecraft-sun angle (remember SSS?)
• Based on JPL trajectory predictions spacecraft
  geometry, GSFC predicted boom would go into
  spacecraft shadow around Dec. 20
• Henry stated that Santa Claus would fix to
  spacecraft before Christmas
• Santa Claus prediction made a believer of ESA IG
  (chief engineer) Massimo Trella (See SOHO later)

30
1980s Ulysses (cont.)

• Prime Mission
• Oct. 1990-Sep. 1995
• Jupiter flyby
• Feb. 8, 1992
• 1st extension
• Oct. 1995-Dec. 2001

Ulysses Orbit Showing North South Solar Polar
Passes
31
1980s Ulysses (cont.)

• 2nd extension
• Jan. 2002-Sep. 2004

Ulysses Orbit Showing North South Solar Polar
Passes
32
1980s Ulysses (cont.)

• 3rd extension
• Oct. 2004-Mar. 2008

Nutation Occurs on this arc
Ulysses Orbit Showing North South Solar Polar
Passes
33
1990s
http//www.cira.colostate.edu/ramm/hillger/GOES-8-
12_image.jpg
GOES (4/13/94)
SOHO (12/2/95)
R
34
1990s GOES-I

• 1st Loral series of geosynchronous weather
  satellites
• Spacecraft acquisition managed by GSFC for NOAA
• Momentum-biased ACS
• Issue Loral safe-hold mode (SHM) was sun-
  pointing zero momentum on thrusters
• Sensitive to earlier satellite losses due to
  control center inexperience
• Not a forgiving SHM, unnecessarily throwing
  away momentum bias
• Could achieve near-zero-momentum state but end up
  in tumble
• GSFC directed a forgiving SHM without thrusters
  which retained momentum bias

35
1990s GOES-I (cont.)

• Once GSFC SHM was implemented, on-orbit test were
  desired
• Center director would not approve in-orbit SHM
  checkout of operational spacecraft
• Henry gets his wish on a dark stormy night at
  230 a.m. on a weekend, when new software upload
  crashed flight computer ground commanded GOES
  to SHM
• SHM was completely successful much more
  stressful than would have ever been permitted for
  checkout

36
1990s SOHO Overview

• Solar and Heliospheric Observatory joint
  international collaboration between ESA NASA
• Designed to study internal structure of sun
• Launched 2 Dec. 1995
• Mass
• Total at launch 1850 kg
• Payload 610 kg
• Dimensions
• Length with solar array deployed 9.5 m

37
1990s SOHO Reaction Wheel (Pre-launch) Story

• 2 of 4 original RWAs failed in less than 1,000
  hours while on ground in Europe
• Time to arrive on-station at Sun-Earth Lagrange
  point (L1) orbit was almost 2,000 hours
• ESA planned only to rebuild 2 failed wheels
  others met spec
• Direction from ESA IG (remember Massimo?) forced
  re-build of all wheels

38
1990s SOHO Orbit

• SOHO slowly orbits L1
• L1 point is about 1.5 million km away from Earth
  in direction of sun
• From this vantage point, SOHO enjoys
  uninterrupted view of sun
• Without rebuild of wheels, SOHO would probably
  have never attained L1
• All wheels still working after more than 11 years
  on station

39
1990s SOHO Loss of Attitude Control in 1998

• Control of spacecraft lost June 1998 due to
  cockpit error
• Spacecraft in flat spin with no communication
  arrays on edge of sun
• Restored 3 months later through efforts of SOHO
  recovery team
• Recovery team had to defrost SOHO
• 2 of 3 onboard gyroscopes never recovered 3rd
  failed, December 1998
• GSFC proposed use of zero-gyro concept, 1st
  developed for IUE spacecraft
• Flight software installed February 1999
• Allowed spacecraft to return to full scientific
  operations

40
Lessons Learned

• Conservation of momentum must hold !
• If at all possible, join with Mother Nature,
  dont fight her
• If at all possible, dont autonomously fire
  thrusters
• In spin stabilized configuration, consider
  nutation as structural resonance of system (SSS
  Ulysses)
• Look for disturbances at nutation frequency, not
  spin frequency
• Safety reliability are opposing requirements
• Be persistent! If something is not right, say so!
  (GOES SHM)
• When accused, the key 1st step is denial (OGO-4)

41
Lessons Learned

• Design philosophies
• Design Backwards Start conceptual design of
  satellite ACS based on what is needed for
  on-orbit normal mode operations then move to
  launch
• Reliability 3 Levels
• No single point failures
• Safe hold whether autonomous or by ground command
  must have ability to reconfigure around failure
  and resume mission after ground intervention
• Fail operational
• In real-time environment, once 1 bad thing
  happens, take time to think about recovery BEFORE
  making another decision!
• Practice omphaloskepsis