Facility for Satellite Formation Flying Control System Verification

1
Facility for Satellite Formation Flying Control
System Verification

• Veres, S.M., Gabriel, S.B., Lincoln, N.K.
• University of Southampton
• (United Kingdom)
• s.m.veres_at_soton.ac.uk, Tel [phone]
• Supported by EPSRC, EADS Astrium, Aero-Astro

2
Outline of talk
Purpose of the facility Modes of operation The
GSF Testing The HIL Testing Software
support Future developments Availability Conclusio
ns
University of Southampton, UK
3
Purpose of the facility

• To complement software based system testing
  with
• Control system hardware testing on the ground as
  much as possible using real sensors and actuators
• To provide hardware-in-the-loop simulations for
  operation in space
• To allow experiments with control algorithms and
  test their long term reliability

University of Southampton, UK
4
Two modes of operation

• A two phase approach is taken to test a
  satellite formation flying system
• Testing of the processors and communication
  electronics with a limited number of sensors and
  actuators (gyros, reaction wheels and air-jets)
  This is called ground-based satellite frames
  testing (GSFT).
• Testing of the actual control algorithms, as they
  would be used in space, using hardware in the
  loop simulations control electronics are real
  but the sensor, actuator and physical dynamics
  are simulated. This is called hardware in the
  loop testing (HILT).

University of Southampton, UK
5
The GSF Testing

• GSFT (ground-based satellite frames testing )
  allows testing of systems software and hardware
  in operation with real components
• solid state gyros (actual)
• accelerometers (actual)
• reaction wheels (actual)
• on board computers (actual)
• communication electronics (actual)
• propulsion substitutes (similar in dynamics)

University of Southampton, UK
6
GSF Testing frames
Satellite frame that can rotate around a central
point (3DOF)
Base disk that glides on a table surface with
very low friction (2DOF)
7
Space ground dynamical differences

• From a dynamical point of view, the essential
  differences are

On the ground
In space
gravitational torques on the ground low
rotation friction on pin point air resistance
more significant Low translational friction on
BTU 5 DOF with constraints
gravity gradient torques in space No rotational
friction No translational friction 6 DOF no
constraints
University of Southampton, UK
8
Testing of satellite autonomous cooperation
Frames in preparation for testing
University of Southampton, UK
9
GSFT facility is capable to test

• 1.1 Suitability of the operating systems on
  onboard hardware
• 1.2 Functioning of the agent based software
  architecture.
• 1.3 Communication devices as they would be used
  in space, so that the system integrity and
  reliability can be assessed.
• 1.4 Some control devices, nearly as they would be
  used in space. Control moment gyros, reaction
  wheels and propulsion substitutes are to be
  tested with real dynamics and not only based on
  manufacturer's descriptions.
• 1.5 The on board sensor and camera system can be
  similar or identical to that used in space. Its
  connection, data throughput and functionality, as
  part of the overall on board digital hardware,
  can be tested.

University of Southampton, UK
10
Metrology infrastructure
University of Southampton, UK
11
Features of the table setup

• The precision-level table rests on a reinforced
  concrete slab supported by a passive vibration
  isolation system to remove ground vibrations.
• Surrounding curtain rails which have been
  tailored for the facility exclude light
  interference from the outside. Internal
  lightening provides homogenous light conditions
  when used.
• The ceiling drape has the potential to imitate a
  portion of the sky with imitation stars.
• Ready made frames are provided that can be used
  to mount custom electronics brought by satellite
  companies or academics. Custom frames can be
  manufactured on site in the EDMC facility of the
  School of Engineering Sciences.

University of Southampton, UK
12
Metrology features
A calibrated 9 camera based observation system
which can monitor the positions and attitudes of
each frame in realtime. The highly sensitive
solid state gyros and the camera based position
and attitude estimates are processed through a
data fusion system which provides realtime
estimates of the 12 dimensional state of each
frame moving on the table.
1 cm accuracy for coarse formation flying
University of Southampton, UK
13
Available GSFT frames

• 3 with model aircraft propeller actuation (8
  bidirectional actuators in each frame)
• 3 frames with air-jet actuators (16
  unidirectional actuators on each frame)
• and 3 frames for future customer use

University of Southampton, UK
14
Hardware in the Loop Testing (HILT)
Onboard computer network with communications (as
in space)
Actuator signals (hardware simulated)
Dynamics of the satellite group (software
simulated)
Sensor signals (hardware simulated)
3D visualisation
University of Southampton, UK
15
HILT Features

• It can use exactly the same control algorithm as
  in space.
• It can use the same control electronics and agent
  based software architecture as GSFT no new
  setup needed
• It can use the same power electronics and
  filtering devices as in space, the actuator
  outputs are replaced by mock-up simulated loads
  with similar dynamics as space certified
  actuators, the sensors are replaced by analogue
  I/O board outputs that are generated through
  simulation software.
• Actual sensor voltages are generated using D/A
  converters and the actuator signal voltages
  sampled by A/D converters

University of Southampton, UK
16
Software support metrology and control
algorithms for the GSFT frames
Feedback linearization based controller
University of Southampton, UK
17
Anticipated use testing of multi-agent systems
on hardware

University of Southampton, UK
18
Future developments

• Precision positioning using laser range finders
• System verification software added
• Compliance with software standards
• Ready made modules with hardware standards
• Etc

University of Southampton, UK
19
Conclusions

• Facility almost ready for industrial use
• Efficient design low cost maintenance great
  value for money in contracted use
• GSFT and HILT are complementary the two
  together address all important issues before
  space use
• Systems are customizable frame size and
  additional electronics can be produced on order
  from us
• We provide dynamical metrology and software
  system for HILT

THANK YOU !
University of Southampton, UK