Diapositive 1

1
What on-board autonomy means for ground
operations An Autonomy Demonstrator conceptual
design
T. Pesquet1, P Grandjean2, AM. Muxi3 1CNES, 18
av. Edouard Belin, 31401 Toulouse cedex 9,
France 2EADS ASTRIUM, 31, avenue des cosmonautes,
31402 Toulouse cedex 4, France 3CS-SI, ZAC de la
Grde Plaine, rue Brindejonc des Moulinais, BP
5872, 31506 Toulouse Cedex 5,France
Abstract The CNES Développement
Exploratoire Autonomie is a coordinated set of
research studies with autonomy as a general
theme conducted within the framework of the
CNESs research and technology plan. This
particular study aims at demonstrating to
spacecraft designers, controllers, and mission
managers the feasibility of extended on-board
autonomy based on state-of-the-art technology.
The goal of this project is to
comprehensively design a ground demonstrator,
simulating an autonomous space mission, the
associated ground control and the external
environment. This technological demonstrator
shall help the space population overcome its
general reluctance to embark autonomy functions
on behalf of the associated risks of losing
direct control on vital processes and prove that
actual on-board autonomy is within reach with an
acceptable reliability level. This study
has been carried out by EADS Astrium and CS-SI,
two of the most prominent members of the French
space industry, under the direction of the CNES,
the French Space Agency. Following studies
will aim at designing the on-board part of the
demonstrator, simulating flying satellites, based
on the system design choices made in the wake of
the presented work, and at developing the ground
segment hereby designed.

• The pattern mission
• The study is conducted with respect to a
  showcase mission, specifically designed so as to
  benefit to a large extent from the operational
  application of autonomy, while at the same time
  very close to current space missions reality.
  This specific mission has a global goal of
  actively monitoring forest fires and active
  volcanoes all over earths surface, using a
  constellation of twelve Low Earth Orbit
  satellites on three distinct orbital planes.
• Each individual satellite performs two main
  tasks,
• on the one hand detecting fires outbursts or
  volcanoes eruptions, using an optical instrument
  with a wide field of view,
• on the other hand acquiring detailed images of
  selected targets with a narrow field instrument.
• The mission combines routine activities
  (global monitoring) and reactive tasks, such as
  the immediate and autonomous acquisition of
  detailed images upon on-board detection of
  starting fires.

• Study framework and methodology
• The first step towards that demonstration
  goal was to define 10 autonomy concepts each
  representing a possible way of implementing
  autonomy that may be associated or operate on
  their own, and combine them in the most efficient
  way, according to the following sequential frame
• What are the most promising autonomy
  concepts combinations? Each combination is
  analyzed, from a system point of view, to
  determine, through analysis or simulation
• Potential impacts on the whole ground segment
  operations, flight-to-ground interfaces, mission
  centers (users-dedicated) and control centers
  (controllers-dedicated) functional scope, as well
  as the operators roles, skills and work
  schedule.
• Expected improvements on the overall cost of
  mission operation, on mission outputs (quality,
  quantity) and the system performances (alarm
  latency).
• What Ground Segment design may support
  these combinations? Different
  architectural solutions are evaluated, so as to
  come up with an operations concept, a
  flight-to-ground functional split, and a ground
  segment functional design.
• How to implement these functions in the
  Demonstrator? Each function is evaluated
  for its feasibility in the demonstrator context,
  taking into account operational software reuse,
  state-of-the-art and prospective technology.

• Introduction
• One of the main drivers when designing and
  building a space system is to cut down the
  associated costs. Upgrading the overall level of
  spacecraft on-board autonomy and thus reducing
  mission operations complexity, while in the same
  time meeting the bound-to-apply safety and
  reliability requirements appears as one of the
  most efficient cost down-cutters available to
  system designers.
• But far beyond the cost factor, on-board
  autonomy enables the space community to consider
  and design unheard-of space missions
• spacecraft constellations that comprise a huge
  number of satellites,
• deep space missions, for which the
  radio-frequency transmission delay is long
  compared to the timing of sequential actions to
  be undertaken by the probe,
• formation flying missions that require a very
  accurate (down to the millimeter or more)
  relative positioning and control of the
  spacecrafts,
• or remote sensing mission in which the events to
  be monitored are transient, and therefore cannot
  be predicted with accuracy (e.g. fire outbursts
  detection), such that a great reactivity is
  required from the system.
• These types of space missions demand such a
  short response time that carrying manual
  operations or even automated ground operations is
  most definitely excluded, and therefore on-board
  autonomy is here, as in a growing number of
  cases, an enabling must.

SDL system modelling and simulation
GEO Relay

• Results
• An analysis of the concepts applied to the
  pattern mission led us to select four concept
  combinations with the highest level of autonomy
  due to some individual concepts complementarity /
  additivity in an operations context. This phase
  was carried out by operations engineers,
  automation experts and mission control designers,
• Simulation based on a SDL model of the system and
  its environment and operation concept led to a
  quantified measure of the immediate returns of
  on-board autonomy based on well defined criteria
  (number of high definition images by target by
  day, elapsed time between a fire outburst and the
  corresponding alarm, TM/TC volume and number of
  antenna contacts, estimated development or
  operations costs),
• A ground segment functional design was built for
  the most promising combinations, with new and
  modified functions clearly identified,
• An implementation trade-off was carried out for
  adapting this functional design to the simulation
  context of the demonstrator.

• Conclusions
• A system study of an autonomous mission,
  identifying precisely and quantitatively the
  advantages of on-board autonomy wrt accurate
  measurement criteria and completed by a risk
  analysis,
• A Ground segment functional design for such an
  autonomous mission, with precise architectural
  schemes and definition of new or modified
  functions,
• A demonstrator preliminary design, based on the
  above ground segment functional design, and
  taking into account technology reuse from
  existing operational products, demonstration
  needs and ease of implementation.
• This will enable us to effectively build
  the demonstrator in the coming months, and help
  future space missions benefit from on-board
  autonomy concepts.