Content uploaded by Vassilis Triantafyllou
Author content
All content in this area was uploaded by Vassilis Triantafyllou on Dec 13, 2012
Content may be subject to copyright.
Aspects of a Collaborative Learning Environment using Distributed Virtual
Environments
C. Bouras,
Computer Engineering and Informatics Dept., Univ. of Patras & Computer Technology Institute
GR-26500 Rion,
Patras, GREECE
bouras@cti.gr
V. Triantafillou
Computer Engineering and Informatics Dept., Univ. of Patras & Computer Technology Institute
GR-26500 Rion,
Patras, GREECE
triantaf@cti.gr
T. Tsiatsos
Computer Engineering and Informatics Dept., Univ. of Patras & Computer Technology Institute
GR-26500 Rion,
Patras, GREECE
tsiatsos@cti.gr
Abstract: A decisive factor for new technologies is always the added value with respect to
the efficiency and capacity of traditional technologies. This also is true when considering the
impact of new technologies in training applications. New types of applications have been
developed along the last few years to incorporate information technology in the learning
environment. The growing need for communication, visualisation and organisation features in
the field of learning and training environments, the e-learning approach, has led to the
application of virtual reality and the use of multi-user real-time communication platforms to
support these needs. This paper presents the first approach of such a system as well as useful
technologies and standards for its implementation.
Introduction
The implementation of an attractive, user-friendly and effective teamwork oriented learning environment, to
offer synchronous and asynchronous training services, includes many technological and pedagogical issues. In
this section a short description of the basic components, issues and requirements of Networked and Learning
Virtual Environments (LVEs) are presented.
A simple Virtual Environment (VE) is a computer system, which generates a 3-D virtual environment, with
which the user can interact and receive real time feedback (Normand et.al. 1999). If multiple are connected and
interact to each other the above definition is extended to multi-user or Shared VE (SVE). A Collaborative VE-
system (CVE) is an SVE aimed at a collaborative task. A Learning Virtual Environment is a CVE that is
designed to offer additional educational tasks such as synchronous and asynchronous learning (Bouras &
Philopoulos & Tsiatsos 2000) and (Bouras & Tsiatsos 2000). A LVE is a set of virtual worlds or a virtual world,
enhanced with educational functionality.
The users are represented by avatars (graphically) that populate the LVE and can be provided with additional
behaviour such as gestures, interaction, movements and sound. Every LVE must comply with a set of
requirements in order to be used widely.
Users are offered a high level of presence through their representation by an avatar of his choice, which
simulates some basic realistic actions, such as gestures and movement, giving them a shared sense of space,
presence and time (Singhal & Zyda 1999). Users are also able to navigate in a 3D shared space in order to access
the content provided, to examine their knowledge, to interact with each other, to exercise their skills and to
receive the information provided. Furthermore the user is informed for the presence of other users (avatars), of
their arrival in the LVE and their leave.
A LVE must also provide their users with many types of interaction in order to enhance the development of users
as autonomous active learners both in the immediate learning context and in the longer term. Two types of
interaction is defined in a LVE:
• Multi-modal user-to-user interaction: chat, voice communication and gestures. This type of interaction is
supported by manipulation of shared 3D objects. Real time applications such as audio communication,
application sharing and whiteboard functionality are important features.
• User-system interaction, which is based on navigational aid and commands that the system provides to the
user for a specific function as well as the manipulation of 3D objects. Furthermore, the users must be able to
insert and change objects in the 3D world, sharing these activities with the other users. This type of
interaction offers the user the capability to customise the total design and outlook of the VEs according to
the needs of their specific themes. Therefore the user-system type of interaction satisfies the need for
customisation.
Although immersive applications are more effective in the use of VR technology, the main feature of educational
VR applications is the interactivity and not the immersion (Youngblut 1998, Sutherland 1968). Moreover, a VR
application, which is designed for educational use should be suitable for widespread use and mature in the part of
the technology. Considering these requirements, immersive VR technology is not mature and it is expensive. On
the other hand desktop VR is more suitable for widespread use regarding the hardware and software
requirements (Youngblut 1998).
A LVE must be scalable to a large number of users in order to support large virtual educational communities.
This set of users can be divided in each virtual world that is a part of a virtual educational community, which is
able to support a maximum number of simultaneous users. The LVE must be able to integrate any digital
material into the platform.
Two other features of the LVEs are consistency and coherence. Consistency is realised by distributing and
synchronising user input as well as user independent behaviour in order to achieve the impression of a single
shared world. Coherence is used with the sense of a uniform structure of the provided services, concerning
mainly the functional and operational characteristics rather than its visual representation in the VEs.
Except the advanced features that a LVE must support, it should also be able to run under a variety of hardware
and software platforms, support different formats/protocols and provide adequate security mechanisms. The
system must offer an easy and complete administration mechanism allowing easy management of training
material and users.
The remainder of this paper is structured as follows. In the next section we describe a European project in the
area of the collaborative learning environments for distance education and our vision for its implementation. We
then present a review of tools, technologies and standards, which are useful for the implementation of such a
project. Finally we present some concluding remarks.
Intelligent Distributed Virtual Training Environment project
INVITE (Intelligent Distributed Virtual Training Environment) [1] is a project in the framework of the
Information Society Technology (IST) Programme of the European Commission. It started in February 2000,
and it will run for almost 3 years.
The main aim of the project is to build a platform for synchronous tele-learning which can be interfaced with
standardised content management and/or instructional management systems. In order to reach this aim the
following objectives have been set:
• Identification of the relevant cognitive and social processes in collaborative learning situation and extraction
of those factors into user requirements.
• Development of an integrated system based on distributed virtual environment technologies, including
intelligent agents’ real-time translation facilities, realistic avatar representation and enhanced interactivity of
avatars.
• Evaluation of the prototype within different learning contexts.
[1] INVITE: Intelligent Distributed Virtual Training Environment project, IST Programme,.
http://invite.fh-joanneum.at/
• Research results on social learning processes within virtual environments.
INVITE provides the user with VR experience so it will be built over a 3D VRML multi-user environment with
smooth movements, extensive textures, 3D icons for manipulation and interaction within the environment and
stereoscopic visualisation options. A virtual model will be used for the construction of the worlds. The user
interface will be enhanced through the integration of 2D video and 3D worlds.
Full body-photo realistic avatars will be used for representing the user since they seem to be more effective when
used in a collaborative environment offering gestures like waving, nodding, bowing, disagreeing etc. The
possibility for using talking avatars (text-to-speech lip synchronisation, driven in real-time from voice) with
movement capabilities and voice driven emotions will be investigated during the project.
In order to offer asynchronous learning services INVITE will integrate available asynchronous learning systems
and support audio/video streaming of available content. Additionally, in order to manage the documents and
other educational material a document repository will be implemented, to facilitate data visualisation and
implementation of structured search engines. A common habit in training sessions is that all participants take
notes on paper copies of the training material and usually the trainer presents his material in slides and allows
access to it.
User interaction will be realised by voice/text chat, and online translation. Users will be supported through the
application of intelligent knowledge based agents thus providing tools for personalised searching and facilitate
the organisation of background information. In (Fig. 1) the main components of a virtual collaborative learning
environment are presented (Triantafillou & Tsiatsos 2000).
Figure 1: Basic Components of Virtual Collaborative Learning environment
Technologies
The range of technologies available for developing collaborative virtual learning environments is more varied
than ever. The exchange and presentation of web based material, 3D content and the transmission of audio/video
streams, of high quality, are now supported by a variety of technologies and tools. Integrated platforms
consisting of a server for the virtual environment and a client for the presentation of the 3D graphics are now
available. Efficient user-system interaction is supported by a variety of intelligent agents. On-line translation
tools for text communication and text-to-speech translation allow multilingual user interaction.
VRML97 (Virtual Reality Modelling Language) and X3D (Extensible 3D) will be used for handling multimedia,
3D objects and shared virtual worlds over the Internet due to their platform independence and their capability to
allow scripts to be embedded adding more functionality in the 3D scene. External user and 3D-scene interaction
will be supported by VRML-EAI (External Authoring Interface).
In order to ensure maximum performance a commercial platform must be used. Such a platform should be open,
allowing easy integration with other platforms and technologies, mature (since the whole project will be based
on the platform), offer full support of VRML and other relevant technologies (HTML, Java) and run on a variety
of systems/servers without major modifications to the code. The most promising platform seems to be the
blaxxun Community Platform that complies with main features such as: full compatibility with VRML,
audio/visual interaction, chat facilities, Java, JavaScript and EAI, 3D rendering and SDK support.
Rapid prototype development of the virtual environments can be supported by Parallel Graphics’ ISB and ISA
which seem to be the most suitable application tools to be used since they support the VRML 97 standard. X3D-
Edit can be used as a graphics file editor for editing, authoring and validation of X3D files. AvatarMe' s
technology (AvatarBooth) will be used for the capturing and creation of photo-realistic avatars.
Educational material (except documents) involves pre-recorded audio and video (streaming media). Most of the
tools that facilitate real time support and electronic conferencing can be used to support communication between
participants in a learning session thus covering the user demand for additional synchronous learning support.
Microsoft Exchange 2000 Conferencing Server and Meeting Point since are open, offer a variety of features,
conform to conferencing standards and can run over various platforms. OnLive platform supports only audio
conferencing but its client can be used as plug-in in a web page. RealNetworks solution can facilitate pre-
recorded streaming multimedia delivery and synchronisation.
User-system interaction in computer based learning systems is provided through intelligent agents (Fabri et. al.
1999). In INVITE a set of functions, like avatar and object handling, mobile communication interaction and
collaboration will be programmed through intelligent agents. Also agents can be reminders, guides, guards and
translators of the users. The use of declarative language implementations provide more advantages than
procedural ones and for that reason the tool to be used should support KQML (Knowledge Query and
Manipulation Language) for communication between the different agents. Java based tools (JAFMAS, IBM
Aglets and AgentBuilder) have an advantage over other tools since they have the ability to run over different
platforms.
HTML and XML will be used for their capability to facilitate efficient document exchange over the Web. XML
offers more capabilities than HTML and can be used to provide metadata and various types of data on the Web.
An efficient database management system will be used in INVITE to manage documents, user profiles and
educational material. Since one of the main objectives of INVITE is to provide a tool that can run over a number
of different platforms this consideration limits our choices concerning the DBMS to be used in INVITE. Oracle'
s solution runs under different platforms and provides a number of important features like security, XML
support, capabilities for producing dynamic XML or HTML documents from SQL queries over the Internet.
On of the most attractive functionality, which must be provided, is the capability for multilingual text
communication and text-to-speech translation. In order to provide such functionality the system will interact with
text-to-text and text-to-speech engines. The most advanced text-to-speech translation tool seems to be Speech
Cube with the capability of running under different platforms allowing a wide variety of languages but the actual
choice of the tool to be applied will depend on its openness and the cost. The most common used system for text-
to-text translation is SYSTRAN, which also can run under a variety of platforms and translate from and to a
number of several languages. Linguatec offers a flexible solution for text-to-text translation that can be used over
the Web.
Standards
The efficient integration of systems relies on the application of world wide accepted standards. A number of
organisations (public, private) are active in this field of technical standards like sub committees of the Joint
Technical Committee of ISO (SC24, SC29, SC34), IETF WebDAV group, and the World Wide Web
Consortium (W3C). In learning technology, interoperability standards and models for learning systems specific
recommendations have been suggested the last few years and a number of committees and projects have worked
towards that direction (IEEE LTSC LOM, CEN/ISSS, CENELEC, ETSI, PROMETEUS).
XML is will be used to facilitate the efficient document exchange over the Web and XSL will support the
creation of virtual XML documents and the presentation of these documents of different media types. The
distribution of interactive multimedia/hypermedia applications in client/server architecture across different
platforms of different types and media will be based on MPEG 7 standard. MPEG-7 will be used to describe the
various types of multimedia information and their relationships using a description language.
Information retrieval will be based on the application of Metadata to identify features shared by different
documents. Metadata for learning material have also been specified such as: LOM, Dublin Core and ARIADNE.
The approach of XMI, which specifies an open interchange model, will support the ability to exchange
programming data between tools, applications and repositories. Learning technology standards will be used for
the organisation and presentation of the learning material as well as a basis for the user interaction specifications.
In this context the IMS specifications defined for easy discovering of data, data sharing over different platforms,
operating systems and tools and ensure reusability will be considered.
User-system interaction will be based on the use of intelligent agents. KQML supports the communication
between the different agents and the implementation of the performatives defined by KQML. KIF (Knowledge
Interchange Format) (Ginsberg 1991) supports the interchange of knowledge among disparate programs.
User-to-user interaction in a LVE will be mainly supported by avatars and chat communication. The integration
of avatars in VE will take into account the specifications for a standard humanoid (H-anim Group). In case these
capabilities are not available electronic conferencing can be used. ITU's T120 and H.323 will be used for audio,
video and data conferencing over the Internet.
Asynchronous learning services to the users will be supported by the delivery of pre-recorded educational
material (audio & video) through the application of real time protocols. (RSVP, RTP RTSP).
The core module in INVITE is the virtual environment since it provides the main functionality and all modules
will use this environment to interact through it. VRML standards for incorporating MPEG-4 technologies (face
and body animation, streamed video and audio etc.) and the recommendations for SQL database access in
VRML will be considered in the development practice of the project.
Communication between the different components of INVITE will lie on the protocols which support client,
server, multicast streaming and network capabilities (VRTP) (Brutzman et. al. 1997), real time interaction
(ISTP) (Waters et. al. 1997), and multi-user participation in virtual worlds (VIP).
Text-to-text and text-to-speech translation standards are set up for lexical resources and language engineering. In
INVITE guidelines from currently advancing initiatives (e.g ISLE and XLT) which produce standards in the
areas of multilingual lexicons, natural interaction, multi-modality, interchange of data among lexical resources
from various translation systems will be used.
Tab. 1 summarises the components of an LVE, useful technologies and tools for its implementation, as well as
standards, protocols and specifications to ensure openness and interoperability.
Components Technologies, Tools Standards, Protocols,
Specifications
Avatars Avatar Studio, Spazz 3D, ICA,
Avatarme' s AvatarBooth H-anim
Real time Audio/Video conferencing Microsoft Exchange 200
Conferencing Server, MeetingPoint H.323, T.120
Streaming Video RealNetworks solution RTP, RTCP, RSVP, RTSP, MPEG
Document Repository XML, MPEG 7, medatada, Oracle 8i LOM, Dublin Core, XML, XMI,
ARIADNE specifications, IMS
specifications
DVE functionality - User Interfaces ISB, ISA, Java, VRML VRML 97, VRML-EAI, VRML-
MPEG4, X3D
Intelligent Agents JAFMAS, IBM Aglets,
AgentBuilder KQML, KIF, OMG and FIPA
specifications
Translation system SYSTRAN, Linguatec' s Personal
Translator 2000, SpeechCube ISLE, XLT
3D Community blaxxun community server VRML97, Java, EAI, X3D
Table 1: Components of a LVE, useful technologies, tools and standards
Conclusion
INVITE aims at the development of a collaborative learning environment for distance education (reflecting real
life collaboration) using distributed virtual environments. INVITE fills in a gap, concerning Networked Virtual
Learning Environments, both in terms of functionality and technology approach. INVITE is a real-time
educational environment, where presence and attendance to the lectures could be made compulsory for inscribed
students with Internet access. The students have the opportunity to participate at the real event of the lecture,
with the ability to raise questions to real professors, or at a specially arranged and recorded event, where the
lecturers are represented by intelligent agents that can be trained to ask commonly asked questions and problems.
INVITE establishes virtual communities with a theme, rules, roles and moderation where useful services can be
employed to facilitate educational procedures. INVITE will provide a tool.
The added value of INVITE is in both the technological and the pedagogical field. INVITE is capable of running
on the average users’ PC and is compatible with standards like VRML enduring openness and portability of the
application. Also, INVITE will facilitate the inclusion of many real world features allowing manipulation of
objects and exchange of information with objects and users. All interfaces between the different modules of the
system are based on mature standards and INVITE is an open system, which can be interfaced with standard
instructional management systems and data representation schemes.
References
Bouras, C., Philopoulos, A., Tsiatsos, T., (2000). Using Multi-user Distributed Virtual Environments in
Education. Proceedings of World Conference of the WWW, Internet and Intranet - WebNet, 2000, (pp. 56-62).
Bouras, C., Tsiatsos, T., (2000). pLVE: Suitable Network Protocol supporting Multi-User Virtual Environments
in Education. Proceedings of International Conference on Information and Communication Technologies for
Education/ED-ICT, 2000, (pp. 73-81).
Brutzman, D., Zyda, M., Watsen, K, Macedonia, M., (1997). Virtual Reality Transfer Protocol (vrtp). Design
Rationale. Presented at Workshops on Enabling Technology: Infrastructure for Collaborative Enterprises (WET
ICE): Sharing a Distributed Virtual Reality, 1997, Massachusetts Institute of Technology, Cambridge
Massachusetts, June 18-20.
Fabri M., Gerhard M., Hobbs J.D., Moore J.D., (1999). Agents for Networked Virtual Learning Environments.
Proceedings of Neties '99 Conference, 1999.
Ginsberg, M., (1991).The Knowledge Interchange Format: The KIF of Death. AAAI Magazine, 12 (3), 57-63.
Normand, V., Babski, C., Benford, S., Bullock, A., Carion, S., Farcet, N., Frecon, E., Harvey, J., Kuijpers, N.,
Magnenat-Thalmann, N., Raupp-Musse, S., Rodden, T., Slater, M., Smith, G., Steed, A., Thalmann, D., Tromp,
J., Usoh, M., Van Liempd, G., Kladias, N., (1999). The COVEN project: exploring applicative, technical and
usage dimensions of collaborative virtual environments. Presence: tele-operators and virtual environments, MIT
Press, 8 (2), 218-236.
Singhal, S., Zyda, M. (1999). Networked Virtual Environments: Design and Implementation. ISBN 0-201-
32557-8, ACM Press.
Sutherland I. E., (1968). A head-mounted three dimensional display. In proceedings of AFIPS Conference, vol.
33, 757-764.
Triantafillou V., Tsiatsos T., (2000). Standardisation & Technology Monitoring Report I - D3.1. INVITE
Document Number: D3.1-CTI-001-Final, Deliverable of the project INVITE (Intelligent Distributed Virtual
Training Environment).
Waters, R.C., Anderson, D.B., Schwenke, D.L., (1997). Design of the Interactive Sharing Transfer Protocol. In
the Proceedings of Sixth IEEE Workshops on Enabling Technologies: Infrastructure for Collaborative
Enterprises, 1997, California: IEEE Computer Society Press, 140-147.
Youngblut, Ch., (1998). Educational Uses of Virtual Reality technology. Institute for Defence Analysis, IDA
Document D-2128.
Acknowledgements
We would like to thank all the INVITE partners of their collaboration and contribution to our work.
