Francisco Luengo

New Goal Selection Scheme for Behavioral Animation of Intelligent Virtual Agents

One of the most challenging tasks in computer graphics and cyberworlds is the realistic animation of the behavior of virtual agents emulating human beings and evolving within virtual 3D worlds. In a previous paper, the authors presented a new, sophisticated behavioral system that allows the agents to take intelligent decisions by themselves [8]. A central issue of this process is the adequate choice of appropriate mechanisms for goal selection. This is actually the aim of the present contribution. In this paper a new scheme for goal selection is described. According to it, the goals priority is calculated as a combination of different agents internal states (given by mathematical functions also described in this paper) and external factors (which will determine the goals feasibility). The architecture of the goal selection module as well as its simulation flow are also analyzed in this paper. Finally, the excellent performance of this new scheme is enlightened by means of a simple yet illustrative example.

Preconditioned Spectral Gradient Method

The spectral gradient method is a nonmonotone gradient method for large-scale unconstrained minimization. We strengthen the algorithm by modifications which globalize the method and present strategies to apply preconditioning techniques. The modified algorithm replaces a condition of uniform positive definitness of the preconditioning matrices, with mild conditions on the search directions. The result is a robust algorithm which is effective on very large problems. Encouraging numerical experiments are presented for a variety of standard test problems, for solving nonlinear Poisson-type equations, an also for finding molecular conformations by distance geometry.

Intelligent agents in virtual worlds

The realistic animation of the behavior of virtual agents emulating human beings and evolving within virtual 3D worlds (a field also known as artificial life) has received great attention during the last few years. The current challenge is to create a behavioral system for the virtual agents so that they behave as realistic as possible. Such a system should provide the agents with the most typical features of human intelligence: perception, recognition, memory, learning, reasoning, etc. Among the large number of different techniques to solve this problem, those based on artificial intelligence (AI) seem to be especially suitable for this purpose. In this paper, we apply several artificial intelligence techniques (such as neural networks, expert systems, fuzzy logic, K-means) to create a sophisticated behavioral system that allows the agents to take intelligent decisions by themselves. The paper describes the behavioral system, its main components and how these AI techniques have been effectively incorporated into the system. In order to show the good performance of our approach, an illustrative example is analyzed.

AI Framework for Decision Modeling in Behavioral Animation of Virtual Avatars

One of the major current issues in Artificial Life is the decision modelingproblem (also known as goal selectionor action selection). Recently, some Artificial Intelligence (AI) techniques have been proposed to tackle this problem. This paper introduces a new based-on-Artificial-Intelligence framework for decision modeling. The framework is applied to generate realistic animations of virtual avatars evolving autonomously within a 3D environment and being able to follow intelligent behavioral patterns from the point of view of a human observer. Two examples of its application to different scenarios are also briefly reported.

A new architecture for simulating the behavior of virtual agents

This paper focuses on modeling the behavior of virtual agents living in a virtual 3D world. Our aim is to apply the most typical human behavior features to our virtual agents so that they behave as realistic as possible. To this end, a new architecture for the behavioral engine that incorporates a number of these typical characteristics of human behavior is introduced. This new proposal allows the virtual agents to interact among them and with the environment in a quite realistic way. The main features of this new architecture, such as perception, knowledge management, motion control and action selection (using internal states, world information, goals, and others) are carefully analyzed in the paper. Finally, some relevant functions (those describing sensations such as tiredness, agents resistance and recovery capacities, happiness and anxiety) and parameters (those determining the vision range or sociability) are also described in the paper.

Interactive 3D Graphics Applications Embedded in Web Pages

As Internet and the World Wide Web (WWW) are becoming more and more popular, there is an increasing demand for new web services and developments. In particular, there is a need of new technologies (often referred to as Web3D) to create and manipulate interactive three-dimensional (3D) environments on the Web. A major problem in this subject is the lack of specialized 3D graphics libraries for Internet, with similar functionalities to those of OpenGL for standalone applications. Although several APIs (Java3D, GJ4JAVA, JOGL, LWJGL, jGL) have been proposed to tackle this issue, none of them has become an official part of core Java package so far. Consequently, it is still hard to build reliable Internet-oriented 3D graphics programs providing the web users with the same realism and interactivity usually found on standalone applications. This paper proposes a based-on-Java framework to develop interactive 3D graphics applications embedded in web pages. Our approach is not based on the development of new programming tools, the emphasis being placed upon the careful selection of already existing technologies (also described in this paper) that provide the best results for our goals along with the optimized design of the workflow. Some applications aimed at showing the potential of our approach are also briefly reported.

GRID Oriented Implementation of Self-organizing Maps for Data Mining in Meteorology

We study the efficiency of different alternatives for a scalable parallel implementation of the self-organizing map (SOM) in the GRID environment of variable resources and communications. In particular, we consider an application of data mining in Meteorology, which involves databases of high-dimensional atmospheric patterns. In this work, we focus in network partitioning alternatives, analyzing their advantages and shortcomings in this framework. As a conclusion we obtain that there is no optimal alternative, and a combination (hybridization) of algorithms is required for a GRID application.

Framework for Simulating the Human Behavior for Intelligent Virtual Agents. Part I: Framework Architecture

This paper is the first in a series of two papers (both included in this volume) describing a new framework for simulating the human behavior for intelligent virtual agents. This first paper focuses on the framework architecture and implementation issues. Firstly, we describe some requirements for such a framework to simulate realistically the human behavior. Then, the framework architecture is discussed. Finally, some strategies concerning the implementation of our framework on single and distributed CPU environments are presented.

Designing an action selection engine for behavioral animation of intelligent virtual agents

This paper presents a new action selection scheme for behavioral animation in computer graphics. This scheme provides a powerful mechanism for the determination of the sequence of actions to be performed by the virtual agents emulating real worlds life. In particular, the present contribution focuses on the description of the system architecture and some implementation issues. Then, the performance of our approach is analyzed by means of a simple yet illustrative example. Finally, some advantages of our scheme and comparison wih previous approaches are also briefly discussed.

A new based-on-artificial-intelligence framework for behavioral animation of virtual actors

The realistic animation of the behavior of virtual actors emulating human beings has been a very dynamic field of research during the last few years. A major goal is to create a behavioral system for the virtual actors so that they behave as realistic as possible. Among the large number of different techniques to solve this problem, those based on artificial intelligence (AI) do represent a powerful (but not well explored yet) approach. In this paper, we present a new framework for behavioral animation of virtual actors. The framework applies several artificial intelligence techniques (neural networks, expert systems, fuzzy logic, K-means) to build a sophisticated behavioral system so that the actors can take intelligent decisions by themselves. The paper describes the general framework, its main components and how these AI techniques have been effectively applied to this purpose. Some programming issues, the main steps of the simulation flow and some illustrative examples are also analyzed in this paper.