A. Sheth et al. (Eds.): ISWC 2008, LNCS 5318, pp. 818–832, 2008.
' Springer-Verlag Berlin Heidelberg 2008
A Semantic Data Grid for Satellite Mission Quality
Analysis
Reuben Wright
1
, Manuel Sánchez-Gestido
1
, Asunción Gómez-Pérez
2
,
María S. Pérez-Hernández
3
, Rafael González-Cabero
2
, and Oscar Corcho
2

1
Deimos Space, Spain
{reuben.wright,manuel.sanchez}@deimos-space.com
2
Departamento de Inteligencia Artificial. Facultad de Informática, UPM, Spain
asun@fi.upm.es, rgonza@delicias.dia.fi.upm.es, ocorcho@fi.upm.es
3
Departamento de Arquitectura y Tecnología de Sistemas Informáticos.
Facultad de Informática, UPM, Spain
mperez@fi.upm.es
Abstract. The combination of Semantic Web and Grid technologies and archi-
tectures eases the development of applications that share heterogeneous re-
sources (data and computing elements) that belong to several organisations. The
Aerospace domain has an extensive and heterogeneous network of facilities and
institutions, with a strong need to share both data and computational resources
for complex processing tasks. One such task is monitoring and data analysis for
Satellite Missions. This paper presents a Semantic Data Grid for satellite mis-
sions, where flexibility, scalability, interoperability, extensibility and efficient
development have been considered the key issues to be addressed.
1 Introduction
Earth Observation is the science of getting data about our planet by placing in orbit a
Hardware/Software element with several observation instruments, whose main goal is
to obtain measurements from the Earth surface or the atmosphere. The instruments on
board the satellite act like cameras that can be programmed to take images of specific
parts of the Earth at predefined times. This data is sent to Ground Stations and then
processed in order to get meaningful scientific information.
Parameters for instrument operations and for the satellite configuration constitute
the Mission Plans issued by the Mission Planning System. These plans are issued
regularly (e.g., on a weekly basis), and can be modified until they are sent to the satel-
lite. Catastrophic events such as earthquakes, volcanic eruptions, and hurricanes are
examples of events that can cause last minute re-planning. These plans and their
modifications are sent to the Flight Operation Segment (FOS), which in turn resends
that information to a Ground Station and from there to the satellite antenna of the
spacecraft. A computer on board the satellite stores the list of MCMD (MacroCom-
mands) that request an instrument or any other part of the satellite to perform an ac-
tion. These include loading a table, triggering an operation and getting internal status

A Semantic Data Grid for Satellite Mission Quality Analysis 819
information. Images from each of the instruments are stored onboard (in the satellite
computer memory) as raw data and when the satellite over-flies the Ground station
that data is sent to the Ground Station antenna (Data downlink). Conversion from the
raw data to higher level “products” (adding identification labels, geo-location data,
etc.) is performed sequentially at the Ground Station and various Payload Data Seg-
ment facilities. Fig. 1 shows the overall scenario. A more detailed explanation of the
whole system can be found in [1].

Fig. 1. General overview of an Earth Observation Satellite system (Envisat)
Among the currently active Earth Observation Satellites we find Envisat, which
monitors the evolution of environmental and climatic changes, and whose data facili-
tates the development of operational and commercial applications. The satellite car-
ries 10 different instruments and is extensively described in [2]. The work presented
in this paper is focused on giving support to this system.
Data circulates within the system as various Plan, MacroCommand and Product
Files, with well-defined structures. There can be a variety of hardware or software
problems that can occur within the process, hence there is a need for the system to be
monitored. QUARC is a system that checks off-line the overall data circulation proc-
ess and in particular the quality of the instrument product files. It checks that the sat-
ellite and instrument have performed successfully the measurements (taking images
of the Earth), that these images have been stored onboard and transmitted as Raw
Data to the Ground station and then processed correctly. QUARC returns reports and
plots, which help in the production of new plans. Additionally, the QUARC system is
designed to assist in decision making when an instrument or the whole system mal-
functions and to detect, in a semi-automated fashion, that something incorrect has
occurred in one part of the product generation or data circulation.
The operational QUARC system is located in a single location (ESA-ESRIN, in It-
aly), which communicates with the archive containing all the products generated from
the beginning of the mission and with all the other facilities. The Data Ingestion
Modules, one per file type, read the files and convert their contents into parameters
that are meaningful to the QUARC data model. The system has been specifically built