Nuevas técnicas para la animación del comportamiento de agentes virtuales autónomos

Abstract

One of the most exciting topics in Computer Graphics is the realistic animation of the behavior of virtual agents. In this field (also known as Artificial Life), the challenge is to create a behavioral system for the virtual agents so that they behave as realistic as possible. To this aim, a number of different techniques have been applied during the last few years. Among them, the most powerful and appealing ones seem to be those based on Artifcial Intelligence (AI) techniques. After all, most of the AI techniques (such as neural networks or expert systems) are based on the idea of reproducing the structure and behavior of the human brain. Consequently, they are optimal candidates to be applied to this goal. This work is focused on describing a general framework for simulating the behavior of virtual actors evolving within a virtual 3D world, and achive realistically to simulate the mental processes of those virtual agents and the application of some well-known AI techniques to this purpose. In particular, we explore some of these techniques and discuss how can they be successfully applied to tackle this issue. For each technique, a discussion about how can it be applied to perform specific tasks is given. We also describe how can some mental processes such as memory, reasonning, recognition, perception and others be realistically simulated by using a carefully chosen combination of these techniques.