From Events to Goals : Supporting Semantic Interaction in Smart Environments
- ISSN: 10848045
- ISBN: 9781424477555
- DOI: 10.1109/ISCC.2010.5546634
Abstract
When we connect smart devices to one another we open up many new possibilities. One interesting possibility is to support high-level semantic interaction without requiring multiple steps on multiple devices. In this paper we investigate how ontologies, runtime task models, Belief-Desire-Intention (BDI) models, and the blackboard architectural pattern may be used to enable semantic interaction for pervasive computing. An initial demonstrator was developed to visualize and manipulate semantic connections between devices in a smart home environment. The demonstrator provides a way for users to physically interact with devices on a high level of semantic abstraction without being bothered with the low-level details.
Author-supplied keywords
From Events to Goals : Supporting Semantic Interaction in Smart Environments
KP / Music PlayerPython for S60Nokia 5800 XpressMusic
KP / Music PlayerPython for S60Nokia N95
KP / Interac on TileArduino-based with Processing /Python (RFID)
TCP Socket over WiFi
TCP Socket over WiFi
Serial over USB
Serial over Bluetooth
KP / Ambient Ligh ng SystemArduino-based
KP / Surround Sound SystemWindows XP with Processing
TCP Socket
Figure 2. An overview of the demonstrator
another connection and now the room is filled with Dries’s
music and colourful lighting effects.
Mark’s roommate Sofia comes back from work and de-
cides she wants to watch a movie on the TV. She seems
somewhat annoyed by the loud music. Mark and Dries do
not want to bother her and they again use the objects to
re-arrange the music stream. Now the music is streamed
to Mark’s portable music player while also playing back at
Dries’s. It is also connected to the ambient lighting system
directly, bypassing the sound system. They both are enjoying
the same music using their own favourite earphones (and
the colourful lighting effects), but without loud music in the
environment. Now Sofia can enjoy her movie without any
disturbing music.”
From this scenario one can see that there are multiple
ways and different levels of interacting with the smart
devices in the environment. There are high-level semantic
interactions with the interaction tile (explore/make/break
connections) and also lower-level interactions with the music
player (play/pause/stop music).
The interaction tile, inspired by Kalanithi and Merrill’s
“Siftables” [17], was designed to explore the connections
and interaction possibilities and manipulation by direct
manipulation, and by making simple spatial arrangements.
The interaction tile visualizes the various connections by
allowing a user to explore which objects are currently
connected, and what connections are possible. Coloured
LED lighting and light dynamics visualize the connections
and connection possibilities between the various devices. By
means of putting devices close to one of the four sides of
the tile, a user can check if there is a connection, and if not,
whether a connection is possible.
A more detailed description of the interaction tile and
the demonstrator is available in an article entitled Semantic
Connections: Exploring and Manipulating Connections in
Smart Spaces, also submitted to this workshop [18]. A visual
Owl:Thing
NFCEventSmartObject
Event NetworkEvent
MediaPlayerEvent
NFCExitEvent
NFCEnterEvent
DisconnectEvent
ConnectEvent
PlayEvent
CueEvent
StopEvent
is-a
is-a
is-a
is-a
is-a
is-ais-ais-a
is-a
is-a
is-ais-a
Figure 3. Ontology indicating rdf:type relationships
overview of the demonstrator can be seen in figure 2.
B. Ontology used in scenario
The ontology used for the prototype was created in
OWL, the Web Ontology Language used to build expressive
ontologies for the Semantic Web.
While developing the ontology, we realized that the most
promising way of describing low-level interactions seemed
to be to describe them in terms of interaction events, which
are traceable, reversible and identifiable. An interaction
event in the smart space consists of an event ID, timestamp
and other related information (e.g. the position of the cube
next to the interaction tile). For the scenario described, we
distinguished between a number of events that can be seen
in figure 3. As a next step we want to determine how we
can use the semantics and expressivity of ontologies to infer
higher-level tasks and goals from these interaction events.
A notable object property used in the ontology is the
connectedTo property, which is both symmetric and
irreflexive. Irreflexive properties are a new feature in OWL 2.
A symmetric property is its own inverse, which means that
if we indicate a connectedTo relationship from device
A to device B, device B will also have a connectedTo
relationship to device A. Another way to think of symmetric
978-1-4244-7755-5/10/$26.00 ©2010 IEEE 1032
Sign up today - FREE
Mendeley saves you time finding and organizing research. Learn more
- All your research in one place
- Add and import papers easily
- Access it anywhere, anytime


