=Paper=
{{Paper
|id=Vol-2262/ekaw-demo-25
|storemode=property
|title=A Proof-of-Concept Implementation of a Semantic Container Management System for Air Traffic Management
|pdfUrl=https://ceur-ws.org/Vol-2262/ekaw-demo-25.pdf
|volume=Vol-2262
|authors=Eduard Gringinger,Christoph Fabianek,Christoph G. Schuetz,Johann Stöbich,Bernd Neumayr,Michael Schrefl
|dblpUrl=https://dblp.org/rec/conf/ekaw/GringingerFSNS18
}}
==A Proof-of-Concept Implementation of a Semantic Container Management System for Air Traffic Management==
https://ceur-ws.org/Vol-2262/ekaw-demo-25.pdf
A Proof-of-Concept Implementation of a
Semantic Container Management System for
Air Traffic Management
Eduard Gringinger1 , Christoph Fabianek1 ,
Christoph G. Schuetz2[0000−0002−0955−8647] ,
Johann Stöbich1,2 , Bernd Neumayr2[0000−0003−1551−172X] , and
Michael Schrefl2[0000−0003−1741−0252]
1
Frequentis AG, Vienna, Austria
firstname.lastname@frequentis.com
2
Johannes Kepler University Linz, Linz, Austria
{schuetz,stoebich,neumayr,schrefl}@dke.uni-linz.ac.at
Abstract. As part of the BEST project – a collaborative research ef-
fort under the Single European Sky ATM Research (SESAR) initiative
– we developed the concept of the semantic container, aiming to facil-
itate information sharing and reuse in air traffic management (ATM)
and air traffic operations, enabled by semantic technologies. In this con-
text, a semantic container is a package of ATM information with an
ontology-based description of the contents. The description serves for
the management of semantic containers: The formal definition of the
contents of a semantic container allows to automatically match existing
semantic containers with an application’s information need. This demon-
stration showcases a proof-of-concept implementation of a system for the
distributed and versioned storage of semantic containers.
Keywords: Semantic Technologies · Distributed Databases · Version
Control · System Wide Information Management.
1 Introduction
Air traffic management (ATM) aims to ensure safe and efficient flight operations
through the timely provisioning of relevant information to stakeholders such
as pilots, air traffic controllers and others. Among the most common types of
information exchanged in ATM, are messages such as Digital Notices to Airmen
(DNOTAMs), which notify stakeholders of important events like runway closures
and malfunctioning airport equipment, and Meteorological Terminal Air Reports
(METARs), which report current weather conditions at airports. Other types of
messages report on weather in regions of airspace, weather forecasts at airports
or in regions of airspace, and significant meteorological events.
Different tasks in ATM and air traffic operations have different information
needs. For example, planning of air traffic flow requires different types of infor-
mation than piloting a flight. Furthermore, piloting a flight from Vienna (VIE)
2 Gringinger et al.
to Frankfurt (FRA) requires different information content-wise than a flight from
Frankfurt (FRA) to Rio de Janeiro (GIG). While pilot briefings for both routes
contain, among other message types, DNOTAMs and METARs, the briefing for
the VIE-FRA flight contains DNOTAMs and METARs relevant for VIE and
FRA airports as well as messages relevant for en-route segments of the flight;
messages for GIG airport will not be part of the VIE-FRA briefing.
The System Wide Information Management (SWIM) concept, which is cur-
rently being developed as part of the Single European Sky ATM Research
(SESAR) initiative, stipulates a service-oriented architecture and standard ex-
change models for sharing ATM information (see [1] for further information).
Hence, information services, which are indexed in the SWIM registry, provide
stakeholders with relevant ATM information in a standard format. In order to
facilitate the packaging and sharing of ATM information inside the SWIM en-
vironment, the BEST project3 developed the semantic container approach. A
semantic container is a package of data items such as DNOTAMs or METARs
with an ontology-based, semantic description of the contents which allows to
match an application’s information need to existing semantic containers. SWIM
information services will then receive semantic containers as input and return
semantic containers as output.
In this demonstration4 , we present a proof-of-concept implementation of a
semantic container management system (SCMS) for the distributed, versioned
storage of semantic containers in ATM. We refer to Deliverables 3.1 [3] and 3.2 [2]
of the BEST project for further information on use case scenarios and the im-
plementation, respectively. The remainder of this paper is organized as follows.
In Sect. 2, we sketch the semantic container approach. In Sect. 3, we present an
overview of the SCMS and sketch integration with the SWIM concept.
2 Semantic Containers
A semantic container has a semantic description, which defines the membership
condition for data items [5]: Every data item, e.g., DNOTAM or METAR, that
fulfills the membership condition belongs to the container. Such a membership
condition has multiple facets – dimensions of content description. With each
facet, a membership condition associates a concept from an ontology. For ex-
ample, a container’s membership condition may associate a concept LOVV with
a spatial facet to indicate that the container incorporates all DNOTAMs rele-
vant for a particular region. Using subsumption reasoning, an information need
expressed as an ontology class can be matched to existing semantic containers.
For example, a membership condition indicating relevance for the LOVV-TS1
route segment is subsumed by a membership condition indicating relevance for
the LOVV region as a whole, which contains that specific route segment.
3
“Achieving the benefits of SWIM by making smart use of semantic technologies”.
For more information, visit the project website: https://project-best.eu/
4
A demonstration video of the implementation is available on the website of the BEST
project: http://project-best.eu/video/best_180604_final.mp4
Semantic Container Management System 3
«XML DB»
Node 1 Data Item Store
REST
Backend
«RDF DB»
Local Registry
Client Browser
Frontend
... «RDF DB»
Global Registry
«XML DB»
Node n Data Item Store
Backend
«RDF DB»
REST Local Registry
Fig. 1. An overview of the semantic container management system
In order to separate the semantics from the storage concerns, i.e., distri-
bution and versioning, we distinguish between logical and physical semantic
containers [4]. Each logical container can be allocated at multiple server nodes,
resulting in multiple physical containers – one for each server where the logical
container is allocated. Furthermore, a physical container maintains multiple data
sets to account for different versions of the physical container when updates oc-
cur. When looking for a semantic container that matches a certain information
need, the logical containers are relevant.
3 System Overview and SWIM Integration
Figure 1 illustrates the architecture of the distributed semantic container man-
agement system (SCMS). For demonstration purposes, we provide a web fron-
tend that allows to create and allocate containers at multiple server nodes. The
frontend accesses via REST API multiple backend nodes at different locations,
e.g., in Vienna, Frankfurt or even on an aircraft. Each backend node maintains
an XML database for the storage of the actual data and an RDF database that
serves as a local registry of physical containers. The different backend nodes
share a global registry of logical containers, which may be (fully) replicated for
increased availability. Thus, the SCMS comprises both a knowledge base for
metadata and a database that contains the actual data items.
Using the presented proof-of-concept prototype, we demonstrate potential
integration of semantic containers from the BEST project into the SWIM net-
work – a BEST-enabled SWIM. In this scenario, an adapted Frequentis SWIM
Registry provides not only information about SWIM services but also about se-
mantic containers. The SCMS is used to define and create containers that are
then visible in the SWIM registry. On an organizational level, the Frequentis
4 Gringinger et al.
SWIM Integration platform (MosaiX) is used to configure the SWIM informa-
tion for the specific SWIM applications. Finally, the information is accessed by
the SWIM applications. To show BEST integration on an application level, we
used an existing prototype – the Integrated Digital Briefing – which was devel-
oped in SESAR 1.
In a BEST-enabled SWIM, semantic containers would facilitate reuse and
caching of information as well as redundant allocation of safety-critical infor-
mation, in addition to tracking changes of the relevant information over time.
Consider, for example, that there are several main routes of air traffic. For each
of these most frequent routes, a semantic container could package the static
information that does not change very often and many flights on that route
require. When collecting the relevant information for a specific flight on such
a route, rather than starting from scratch, the application could start with the
already existing container. An application could then customize the container by
creating a new container, based on the existing container, that adds additional
information relevant for the specific flight. Redundant allocation of semantic con-
tainers is important for safety reasons. Important information could be stored
in multiple locations, including the aircraft. Tracking changes in the actual data
set of a physical container over time is important to ensure auditability in case of
accidents. That way, the version of a physical container in a particular location
at a particular point in time can be reconstructed.
Acknowledgments. This project has received funding from the SESAR Joint
Undertaking under grant agreement No 699298 under the European Union’s
Horizon 2020 research and innovation program. The views expressed in this
paper are those of the authors.
References
1. EUROCONTROL: System wide information management (2018), https://www.
eurocontrol.int/swim (Accessed: 26 October 2018)
2. Gringinger, E., Fabianek, C., Schuetz, C.G.: D3.2 – Prototype SWIM-Enabled Ap-
plications. Tech. rep., BEST Consortium (2018), http://project-best.eu/downloads/
3. Gringinger, E., Neumayr, B., Fabianek, C., Savulov, A.: D3.1 – Use Case Scenarios.
Tech. rep., BEST Consortium (2018), http://project-best.eu/downloads/
4. Gringinger, E., Schuetz, C., Neumayr, B., Schrefl, M., Wilson, S.: Towards a value-
added information layer for SWIM: The semantic container approach. In: Proc. 18th
ICNS Conference. pp. 3G1–1–3G1–14 (2018)
5. Neumayr, B., Gringinger, E., Schuetz, C.G., Schrefl, M., Wilson, S., Vennesland, A.:
Semantic data containers for realizing the full potential of system wide information
management. In: Proc. 36th DASC (2017)