|
February 2003 |
International Journal of
University
of Wolverhampton UK in collaboration with the European Simulation Society (EUROSIS) |
|||
|
Chief Editors Editorial
Board & Aims Previous issue: Copyright University of Wolverhampton and EUROSIS 2003 ISSN 1477-2043 The Editors, The University of Wolverhampton and EUROSIS do
not accept responsibility for errors arising from papers
appearing in IJIGS. Official
University of Wolverhampton disclaimer. |
Abstracts Coordinating piece
movements in a semi-concurrent abstract board game T. PannérecIn semi-concurrent games, each player simultaneously moves a set of pieces, the object of the game being to coordinate these movements to maximise the winning chances. In this paper, we present such a game, discuss the problem it poses and report the use of our MARECHAL framework to model the tactical and strategic expertises. The results show that our AI opponent can globally play at an experienced human player level. Dynamic character animations S. M. GrünvogelFor
creating real-time animations of 3D characters we introduce an animation
engine for the dynamic creation of motions with motion models. Each motion
models represents a small task like walk or wave and has its
own set of parameters controlling the specific characteristics of a motion.
The style and the characteristics of motion models can be changed in
realtime. For this purpose pre-produced animations are combined and changed
with clip operators. The animations of several motion models can be combined
to play different motion simultaneously. To create the combined motions, the
same clip operators are used. Improved opponent intelligence trough offline learningP. Spronck, I. Spinkhuizen and E. Postma Artificially intelligent opponents in commercial
computer games are almost exclusively controlled by manually designed
scripts. With increasing game complexity, the scripts tend to become quite
complex too. As a consequence they often contain “holes” that can be
exploited by the human player. The research question addressed in this paper
reads: How can evolutionary learning techniques be applied to improve the
quality of opponent intelligence in commercial computer games? We study the
offline application of evolutionary learning to generate neural-network
controlled opponents for a complex strategy game called PICOVERSE. The results show that the evolved opponents outperform a
manually-scripted opponent. In addition, it is shown that evolved opponents
are capable of identifying and exploiting holes in a scripted opponent and
exhibiting original tactical behaviour. We conclude that evolutionary
learning is an effective tool to improve the quality of opponent intelligence
in commercial computer games. Creating socially interactive no-player characters: The µ-SIV systemB. McNamee and P. Cunningham A number of recent,
highly successful games have shown that there is a demand for the
personalities, moods, and relationships of Non Player Characters' (NPCs) to
be made the focus of game-play. In order for this shift of focus to take
place, agent architectures used to create NPCs must be augmented with models
of these aspects of characters’ personae. These models must then be used to
drive characters' behaviour. This paper will present the µ-SIC system, a
component of an intelligent agent architecture designed for the creation of
NPCs for computer games. µ-SIC uses a number of quantitative psychological
models to simulate characters’ personalities, moods and relationships. The
values of these models are used as inputs to an Artificial Neural Network
(ANN) which drives characters’ social behaviour. The execution kernel of RC++: RETE*, a faster RETE with TREAT as a special caseI. Wright and J.
Marshall Some behaviours of computer game agents can be
naturally expressed as collections of rules and knowledge bases. General
purpose rule-based languages provide high-level constructs for expressing
complex conditional behaviour. We examine the runtime kernel of RC++, a
rule-based language developed for game AI, to explore the costs associated
with adopting general-purpose, rule-based approaches for computer game
production. The kernel of RC++ is the RETE* algorithm, an extension of the
RETE algorithm with better time characteristics, but also able to exhibit the
beneficial properties of TREAT (a low memory cost alternative to RETE) when
required. RETE* achieves this functionality and performance by employing (i)
asymmetric deletion, (ii) dual tokens, and (iii) a dynamic beta-memory cut
mechanism. The dynamic beta cut allows the RETE/TREAT trade-off to be
exploited by users. Theoretical and empirical performance comparisons for
RETE, TREAT and RETE* are provided. The implications for the utility of
rule-based programming for the computer games industry is discussed, and we
conclude that there is still some way to go before rule-based programming can
be employed in the game-making process. Video games in educationK. Squire Computer and video games are a maturing medium and industry and have
caught the attention of scholars across a variety of disciplines. By and
large, computer and video games have been ignored by educators. When
educators have discussed games, they have focused on the social consequences
of game play, ignoring important educational potentials of gaming. This paper
examines the history of games in educational research, and argues that the
cognitive potential of games have been largely ignored by educators.
Contemporary developments in gaming, particularly interactive stories,
digital authoring tools, and collaborative worlds, suggest powerful new
opportunities for educational media. 4
|
|||
