The Development of Intelligent Autonomous Software Agents for Virtual Environments and other Areas of Telematics

Virtual Environments (VE), often also called Virtual Realities (VR), are a special case of "cyberspaces", which enable the user, who usually wears a head-mounted display and a data glove, and sometimes also a data suit, to "totally immerse" as well as navigate and manipulate in a virtual environment. The big potential of VEs is illustrated by numerous conferences, publications and already also by products.

An important requirement for the successful use of VEs as well as of textbased cyberspaces (information networks, MUDs, etc.) would be intelligent autonomous software agents, which would enable the user to find e.g. a certain place or a certain information, or data about some interesting activity, or to construct complex VEs quickly and efficiently. Moreover, many cyberspaces, for example in the field of "edutainment", would gain real attractivity only when intelligent partners could interact with the user.

For these reasons, the development of intelligent autonomous agents for VEs and other cyberspaces becomes increasingly important. At the beginning of the 1990s, however, this research was carried out almost exclusively in the United States. Consequently, it had been warned in Europe that, in the field of VEs, "inaction is the high risk policy, with possible losses of vast potential markets."

A necessary basis of principled research in this area, this project resulted in an overview of autonomous agents research and the setting up of connections to research and development institutions that were already working in this field in the USA or started working in this area in Europe.

Project Leader:

Univ.-Prof.Dr.Robert Trappl

Researchers:

Dipl.-Ing.Dr.Paolo Petta
Dipl.-Ing.Dr.Bernhard Pfahringer
and students

Project Period:

June 1994 - 1996

Sponsor:

Austrian Federal Ministry for Science, Research, and the Arts

Publications:

Trappl R.: Il Ruolo dell' Intelligenza Artificiale nella Realtà Virtuale, Virtual: Mensile di Realtà Virtuale e Immagini di Sintesi, Milano (in press), 1995.

Trappl R.: Soziale Aspekte im Cyberspace, Österreichisches Forschungsinstitut für Artificial Intelligence, Wien, TR-93-32, 1993.

Trappl R., Petta P.: What Governs Autonomous Agents?, in Magnenat-Thalmann N. and Thalmann D.(eds.), Proceedings Computer Animation '95, April 19-21, Geneva, Switzerland, IEEE Computer Society, Los Alamitos, CA, pp.1-10, 1995.

Petta P., Trappl R.: On the Cognition of Synthetic Characters. in Trappl R.(ed.), Cybernetics and Systems '96, Proceedings of the 13th European Meeting on Cybernetics and Systems Research. Austrian Society for Cybernetic Studies, Vienna, pp.1165-1170, 1996.

Petta P., Trappl R.: Why to Create Personalities for Synthetic Actors. in Trappl R. & Petta P.(eds.), Creating Personalities for Synthetic Actors, pp.1-8, Springer, Heidelberg/New York, 1997.

Petta P., Trappl R.: Personalities for Synthetic Actors: Current Issues and Some Perspectives, in Trappl R. & Petta P.(eds.), Creating Personalities for Synthetic Actors, pp.209-218, Springer, Heidelberg/New York, 1997.

Trappl R., Petta P.(eds.): Creating Personalities for Synthetic Actors: Towards Autonomous Personality Agents, Springer Lecture Notes in Artificial Intelligence 1195, Springer Verlag, Heidelberg/New York, 1997.