Enhancing Believability of Virtual Soccer Players: Application of a BDI-Model with Emotions and Trust - Developing Concepts in Applied Intelligence - page 118

Enhancing Believability of Virtual Soccer Players:
Application of a BDI-Model with Emotions and Trust
Tibor Bosse and Daniel Höhle
Vrije Universiteit Amsterdam, Department of Artificial Intelligence
de Boelelaan 1081a, 1081 HV Amsterdam, The Netherlands
[e-mail], [e-mail]
Abstract. Despite significant progress in the development of virtual soccer
games, the affective behavior of virtual soccer players, both in software and
hardware applications, is still limited. To improve this situation, this paper
presents a generic model for decision making of virtual agents in relation to
emotions and trust. The model takes the BDI framework as point of departure,
and extends this with mechanisms to represent the dynamics of emotions and
trust. After testing the model by means of simulation experiments, it has been
incorporated into the virtual agents within the RoboCup 2D soccer
environment. A preliminary evaluation pointed out that the model successfully
enhances believability of virtual soccer players.
Keywords: soccer simulation, cognitive modeling, affective computing, trust.
1 Introduction
Association football, commonly known as soccer, has been the world's most popular
sport for many decades. Thanks to the rapid developments in the video game industry
as well as scientific areas such as Computer Science and Artificial Intelligence,
nowadays, soccer games are not only played in the real world, but also in the virtual
world. Overall, the purpose of virtual soccer games (or soccer simulations) is three-
fold. First and foremost, virtual soccer games are played for entertainment purposes;
in these cases, one or multiple human users control the soccer players in a virtual
environment, with the goal to defeat each other or a computer opponent [12]. Second,
soccer simulations are increasingly being used for the purpose of analysis, for
instance by managers and coaches of real soccer teams; here, the idea is that the
virtual environment provides an easy framework to test various strategies that cannot
easily be tested in the real world. For example, in case a coach intends to investigate
what might happen if his team uses a particular tactical strategy in a particular game,
he could run a simulation of this game, where each of the simulated players behaves
according to this strategy. Although this area is yet to be explored in much more
detail, some anecdotical cases have been reported in which soccer simulations were
able to make accurate predictions (e.g., [15, 17]). Third, virtual soccer games have
proved to be an excellent test bed for the application of a variety of AI techniques,
including (real-time) multi-agent planning, multi-agent communication, behavior
modeling, and learning. These challenges have been the main motivation for the
foundation of the RoboCup competition in 1997 [10].