European Space Agency

1
European Space Agency
The ESA Mission experience
ESA Ground Operation System (EGOS) 2004
Workshop June 15-16 2004
M Jones, A Ercolani, F Delhaise, D. Guerrucci,,
G W Kerr, M. Merri, M Pignede, D. Ponz, V.
Reggestad, G di Girolamo

2
Presentation Roadmap

• What OPS-GD does
• Effective software reuse and the delta approach
• Survey of mission projects done by GD
• Planetary
• Earth Explorer
• Low cost Metop-1 and Smart-1
• S2K Migration (XMM-Newton)
• Conclusions

3
OPS-GD Mission Data Systems Division
Reuse of Software a Simple Schematic
4
Missions implemented by OPS-GD using SCOS-2000
and NCTRS
Table 1 Launched Missions
Table 2 MCS under Development
5
Effective Reuse the Delta Approach

• Old practice complete requirements were written
  for each new data system even if significant
  reuse was envisaged
• Mismatches to capabilities of reused software
• Wasted effort
• Solution delta requirements document covering
  only new or changed requirements
• Typically for a MCS, the requirements will be a
  delta to the SCOS-2000 requirements
• Can be extended to rest of the software life
  cycle, producing deltas to Architectural Design
  Document, software test documents etc.

6
Mission Families

• A mission family is a set of missions based on
  the same type of spacecraft platform or bus
• There will be typically a common set of TM/TC
  data types, structures and services
• Differences will arise from the payloads and the
  detailed mission definition.
• Can expect to use the deltatechnique, but this
  time applied with respect to the mission family
  kernel
• Even great benefits for simulator development
  all the platform models are shared

7
Example Delta Requirements between Missions

• Inner circle
• SCOS-2000 kernel Requirements
• Second circle
• Requirements mission family kernel
• Normally first mission M1
• Outer circle
• Requirements mission M2,
• Where M1 and M2 are two
• members of a mission family

M2
Family Kernel
SCOS-2000 Requirements
8
Planetary Mission Family - Overview

• Currently comprises Mars-Express, Rosetta and
  Venus Express missions
• MCS was first developed for Rosetta a
  comet-chaser mission
• Then adapted for Mars-Express
• Because of launcher problems, Mars-Express was
  launched first (in June 2003).
• Preparation of Venus Express currently in
  progress
• Family reuse approach also successfully applied
  to the simulator for all three missions

9
Planetary Mission Family

• Mission timeline management in deep space
• File Transfer Protocol
• Combined Processing of real time and offline
  telemetry - replay functionality
• Packet displays
• Data Disposition System
• Improvement of the Filing Performance
  (Mars-Express)
• Long term Archive Concept (Mars-Express)
• New Archive Split Function

10
Earth Explorer Mission Family - Overview

• Comprises three missions CRYOSAT, GOCE and
  Aeolus
• Cryosat launch foreseen end 2004, GOCE 3Q/2006,
  Aeolus late 2007
• Cryosat is the first mission in the family and
  launch preparation is on-going
• GOCE MCS development is in progress taking
  effective advantage from the Cryosat MCS
  development and test activities

11
The EE Mission Family Delta Approach Starts

• A high level of harmonisation has been achieved
  in the on-board design of Cryosat and GOCE in all
  respects affecting the ground segment
• Helped by ground segment teams involvement in
  the design of the satellite TM/TC implementation
• Delta MCS development approach

12
The EE Mission Family - Requirements Sets Today

• Requirements defined in S2K not used by any EE
  Missions today, but not contradicting EE
  Missions, therefore included in EE Kernel
• Requirements in EE Kernel not common to all EE
  missions, but not contradicting any EE Mission.
• Mission Specific requirements
• Requirements implemented by mission and common to
  all EE Missions (candidate for integration in EE
  Kernel)

13
Metop1 - Introduction

• An ESOC third party mission
• Launch and Early Orbit phase of Eumetsats
  Metop1 polar orbiting weather satellite carried
  out by ESOC
• Challenges
• Complex spacecraft bus (same as ENVISAT, which
  is Europes largest EO mission)
• Very low budget (about 1000 KEURO set aside for
  all MCS life cycle costs, including development,
  preparation support and software support during
  the LEOP)

14
Metop1 addressing the challenge of low-cost

• Cost constraints were addressed by
• early use of two ESOC software contractor firms
  to promote knowledge acquisition
• maximize software reuse, primarily of the the
  SCOS-2000 kernel
• limiting the requirements to what is strictly
  needed to operate safely a LEOP
• ensuring that the requirements specification was
  very specific without TBDs and agreed formally by
  all parties so as to minimize the risk of
  requirements creep
• contracting via a competitive tender involving
  knowledgeable suppliers

15
SMART-1

• SMART-1 as the first of a class of 'Small
  Missions for Advanced Research in Technology
• SMART-1 demonstrates electric ion propulsion
  technology by using it to reach the Moon.
• SMART missions are low-cost (ca. 100 Meuro)
• Meant that low-cost solution had to be developed
  for ground segment and in particular MCS.
• Launched in September 2003, with moon orbiting in
  early 2005

16
SMART-1 Software Reuse

• Another low cost mission
• As well as SCOS-2000, Smart-1 reused significant
  sub-systems from other missions
• New code developed for SMART-1 accounted for only
  6 of the total code

SMART-1 Software Reuse in units of number of
modules (files) from each contributing
project Reuse 93.6 Leverage 16
Leverage Productivity Gain
17
XMM-Newton MCS and Science Operations Centre
(SOC) Migration Project

• XMM-Newton is currently migrating the SCOS-1
  based control system, running on VMS operating
  system on Alpha platform to SCOS-2000/Sun
  Solaris. Systems concerned
• Mission Operations Centre (MOC)
• Science Operations Centre (SOC).
• Main drivers for the project are
• ensuring functionality of old system is supported
• Improving long term maintainability in hardware
  and software,
• Significantly reducing OM costs.

18
XMM-Newton Migration

• maximize synergies between MOC and SOC
  developments, by sharing code of common
  subsystems,
• Extensive re-use of SCOS-2000
• Porting of mission specific subsystems like
  Payload Monitoring Subsystem (PMS), by developing
  a middleware layer to encapsulate operating
  system and SCOS-2000 infrastructure interfaces
• Implemented for a (small) fraction of the cost of
  the original MOC and SOC developments

19
Conclusions - 1

• Cost trend the following chart shows
• development costs for two families (Planetary,
  Earth Explorer)
• evident decrease from the family kernel to the
  subsequent missions in each family

20
Conclusions -2

• Extensive reuse of mission control software at
  ESOC
• Several major types of reuse described
• Infrastructure
• Mission family
• From infrastructure and other missions (Smart-1
  case)
• All types of reuse give major cost savings
• Significant transfer of capability from Mission
  developments to the infrastructure
• A two-way exchange
• Reuse has ensured the steady fall in MCS
  implementation costs