=Paper=
{{Paper
|id=Vol-260/paper-17
|storemode=property
|title=How to build a Context-aware Architecture for Ubiquitous VR
|pdfUrl=https://ceur-ws.org/Vol-260/paper17.pdf
|volume=Vol-260
|dblpUrl=https://dblp.org/rec/conf/isuvr/OhW07a
}}
==How to build a Context-aware Architecture for Ubiquitous VR==
https://ceur-ws.org/Vol-260/paper17.pdf
International Symposium on Ubiquitous VR 2007 1
How to build a Context-aware Architecture for
Ubiquitous VR
Yoosoo Oh and Woontack Woo
seamless connection among users and real/virtual
Abstract— There are no general context-aware architectures to environments. Previous context-aware architectures are not
be useful which could be applied to full life cycle in ubiquitous VR, proper to use and apply to real and virtual environments. Those
even though many context-aware architectures or middleware are should be considered more about connection among
proposed. In this paper, we propose a general approach of a
context-aware architecture for ubiquitous virtual reality. We
domain-free entities. Thus, more general approach to
would discuss how to improve the intelligence ability in a context-aware architecture should be developed. In this paper,
context-aware architecture. we propose a general approach of a context-aware architecture
for ubiquitous virtual reality. Also, we want to discuss some
Index Terms—Context Interpretation, Context Learning, issues which are how to improve the intelligence ability in a
Context Reasoning, Social Relationship, Ubiquitous VR context-aware architecture.
There are no general context-aware architectures to be useful
which could be applied to full life cycle in ubiquitous VR, even
I. INTRODUCTION though many context-aware architectures or middleware are
U biquitous VR is a concept of creating ubiquitous
VR(virtual reality) environments which make VR
pervasive into our daily lives and ubiquitous by allowing VR to
proposed. Also, it’s hard to apply the working architectures or
middleware to a new domain that is not mentioned in their
original works. Thus, we should discuss how to build
meet UbiComp(ubiquitous computing) [1]. To make seamless standardized and generalized context-aware architectures for
connection between human beings and VR and UbiComp various environments.
environments, we need to have an infrastructure which is
specified by sensors and services for the collaboration of those
environments. In different domains like human, real and virtual II. CONTEXT-AWARE ARCHITECTURE: UCAM
environment, a context-aware architecture for Ubiquitous VR
is necessary to be established. A. UCAM architecture
There are many previous works related to a context-aware UCAM is a Unified Context-aware Application Model for
architecture. CAMidO is a Context-Aware Middleware based seamless human-content-environment interaction in ubiquitous
on an Ontology meta-model [2]. CAMidO Compiler generates computing environment [5]. Figure 1 shows UCAM
Inference rule file and controller source code. However, architecture. For the general approach, we modified previous
CAMidO has the sophisticated path in architecture and has the UCAM. Our proposed UCAM particularly supports
weak relationship among context sources. JCAF is the Java intelligence like context reasoning, learning, interpretation, and
Context Awareness Framework [3]. JCAF cooperates with social relationship management. The proposed UCAM
context services and supports for adding, deleting, modifying, architecture used context model which describes context as a
or varying functionality, capacity, or platform at runtime. JCAF form of 5W1H (Who, What, Where, When, How, and Why) [5].
should solve synchronizing data used in many systems. The context model also maintains context history as context
SOCAM is a Service-Oriented Context-Aware Middleware [4]. repository.
SOCAM supports context reasoning and knowledge sharing The proposed architecture can improve the awareness ability
process by using common context model. SOCAM depends on of context because the architecture can interpret, reason out,
OWL, and so it’s not easy to be used widely. Besides these and learn contexts. The architecture can connect any tool for
architectures, there are many related works. the awareness ability as an add-on feature. Also, the
However there is still a need for a generic solution for architecture manages social relationship among human, real
objects, and virtual objects. For these aspects, our architecture
This work was supported by the IT R&D program of MIC/IITA. is generalized as an open source code which is a library for
[2005-S-069-02, Development of Technologies for Unifying and Fusing
Context]. basic features. An application developer can simply use our
Yoosoo Oh is with the Gwangju Institute of Science and Technology,
Gwangju 500-712, S.Korea (e-mail: yoh@gist.ac.kr).
Woontack Woo is with the Gwangju Institute of Science and Technology,
Gwangju 500-712, S.Korea (corresponding author to provide phone:
+82-62-970-2226; fax: +92-62-970-2204; e-mail: wwoo@gist.ac.kr).
International Symposium on Ubiquitous VR 2007 2
architecture by inheriting the library code. parameter mapping (Domain Change). Context interpreter
maps elements in context format to each application domain,
Application and one application to another application. It requires
Context
knowledge about both context format and application
UCAM (Service) Command
Model parameters for interpretation.
(Context
Service Provider Repository)
SCC 5) Social Relationship Management
Social Relationship Management in UCAM determines
Interpreter UCC Context Manager relationships among sensors and services which participating in
IC/FC
ubiquitous VR environment (human, real/virtual env.). To
manage social relationship among entities in ubiquitous VR
Context Integrator FC
environment, those entities (users, services and sensors) should
PC/FC collaborate each another in the created community. This
Communicator process is done by Communicator in UCAM which takes
Network
charge of network configuration and sends messages to
Fi
gure 1. UCAM service part. (PC: Preliminary Context, IC: Integrated Context, community members.
FC: Final Context, UCC: User Conditional Context, SCC: Service Conditional
Context) III. DISCUSSION
The proposed architecture can improve the awareness ability
B. Core technology and extend to diverse application domains. Also, the
1) Context Model architecture, UCAM, is generalized as a library in order to be
Context Model in UCAM is defined a set of user contexts used easily by application developers. However we still have
that supports ontological structure and reasoning operations. It many issues to be discussed. First issue is how to predict user’s
has to know what kinds of operations are needed in each behavior and connect to the other objects in real/virtual
component in UCAM. It is related to all components in UCAM environment. Basically, UCAM can support the reasoning
and generates and saves contexts. aspect, not prediction for user’s behavior. For more
generalization, this issue should be considered. Second issue is
2) Context Reasoning full life cycle support for context-aware applications. Adding,
Context Reasoning in UCAM is a logical process which deleting, replacing, modifying, and debugging entities should
extracts high-level(implicit) context from low-level(explicit) be supported. These functionalities are already supported by
context and maintains context consistency in processing UCAM, but need to be improved for ubiquitous VR
context modeling. For context reasoning, it is necessary to environment. Last issue is how to evaluate a context-aware
various kinds of sensors and information sources. Context architecture. There are good evaluation methods in HCI field.
reasoning process reuses context (context history) and adapts Thus, those methods should be applied to our UCAM with
reconfigurable rules to the diverse environment. It is related to some evaluation factors, such as architecture structure,
sensor part of UCAM and Context Integrator in UAM service. scalability, accountability, reconfiguration, reusability,
debugging, openness, flexibility, etc. These evaluation factors
3) Context Learning could be useful to verify the general approach and intelligence
Whether the context-aware application obtains its context ability of our UCAM.
information from environmental sensors, user input, VR
applications, or some combination of these, it must perform a REFERENCES
good deal of processing on the long-term context in order to be [1] S.Kim, Y.Suh, Y.Lee, W.Woo, “Toward ubiquitous VR: When VR Meets
ubiComp,” ISUVR 2006, pp. 1-4, 2006.
able to accurately evaluate the state of the environment, the [2] Nabiha Belhanafi Behlouli, Chantal Taconet and Guy Bernard, “An
intensions of the user, and avatar’s intelligent reaction. Thus, Architecture for supporting Development and Execution of
Context Learning in UCAM is required to context-aware Context-Aware Component applications,” In IEEE International
Conference on Pervasive Services, ICPS 2006, Lyon, France, 26-29 June,
applications. Context learning is related to Context Manager in
2006.
UCAM, and uses classified learning mechanisms according to [3] Jakob E. Bardram, “The Java Context Awareness Framework (JCAF) - A
the characteristics of them. Service Infrastructure and Programming Framework for Context-Aware
Applications,” In Proceedings of the 3rd International Conference on
Pervasive Computing (Pervasive 2005), 2005.
4) Context Interpretation [4] Tao Gu, Hung Keng Pung, Da Qing Zhang, “Toward an OSGi-Based
Context Interpretation in UCAM is the way the entity Infrastructure for Context-Aware Applications,” IEEE Pervasive
generates conditional context for performing certain task, and Computing, vol.03, no.4, pp.66-74, Oct-Dec, 2004.
[5] D.Hong, C.Shin, Sj.Oh, W.Woo, “A New Paradigm for User Interaction
the way the entity presents current context to the in Ubiquitous Computing Environment,” ISUVR 2006, pp. 41-44, 2006.
entity-readable way. It generates conditional context from
user's direct input (Feedback) and do context-to-service