BDI agents for game development

Authors:
Arran Bartish;Charles Thevathayan
Affiliations:
Royal Melbourne Institute of Technology, Melbourne, Victoria, Australia;Royal Melbourne Institute of Technology, Melbourne, Victoria, Australia
Venue:
Proceedings of the first international joint conference on Autonomous agents and multiagent systems: part 2
Year:
2002

Citing 1
Cited 3

Software Engineering: A Practitioner's Approach

Software Engineering: A Practitioner's Approach

CogBox: combined artificial intelligence methodologies to achieve a semi-realistic agent in serious games

ICAISC'10 Proceedings of the 10th international conference on Artifical intelligence and soft computing: Part II
Unreal goal bots: conceptual design of a reusable interface

Agents for games and simulations II
Improving behavior of computer game bots using fictitious play

International Journal of Automation and Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The study of game related topics has long been the subject of research and development, however game Artificial Intelligence (AI) predominantly uses classical AI simulation techniques. Older computational models such as state machines continue to be used as the back bone for game AI development even though alternative models are available. In this research, we have attempted to shed some light on how the choice of implementation may affect the important factors such as performance, and complexity as the games scales up. Our findings show that agents and finite state machines have their relative merits. We have shown that complexity measured as a function of the number of behaviours, was linear for agents and quadratic for FSM. Though run-time performance is comparable for a small number of entities, it degrades at a higher rate for agents.