Development of intelligent virtual agents is a complex task. One of the im- portant sub-tasks is creation of behaviors. It is highly desirable to test behaviors of intelligent virtual agents in an actual virtual world, not in a simplified sub- stitute. Over the past several years the Pogamut platform has been developed, which allows intelligent virtual agent behaviors to be tested in several gaming worlds, but none combines complexity, dynamicity and extensibility. It was de- cided to fill the gap by creating the EmohawkVille virtual world. EmohawkVille is a first-person virtual world in a day-to-day life setting. The purpose of this thesis is to explain why is EmohawkVille needed and to document features, de- sign and development of EmohawkVille and its Pogamut module. The thesis also presents results of a case study that confirms the EmohawkVille's suitability for experiments concerning intelligent virtual agent behaviors.

Affordable and relatively accurate eye trackers are becoming more accessible for the mainstream user, consequently opening up new opportunities for utilizing eye gaze in interactive real-time applications. When extending gaze data from an eye tracker into real-time rendered 3D virtual environments, there are often techniques applied to correlate the gaze point to a 3D object. Techniques which can be referred to gaze-to-object mapping (GTOM) methods. One featured GTOM method, region-bound Interactors, provided by the eye tracking company Tobii AB, was extended in this thesis by utilizing a custom G-buffer in order to extract accurate shapes of individual objects in screen space. This adapted method was compared against two methods based on ray casting in a controlled 3D test environment consisting of three types of scenarios with varying conditions. Data was collected from a user study with 31 participants to assess if the Interactors method could be viable in 3D virtual environments. The extended method showed some promising result in relation to the other methods, but further work needs to be done to establish whether the method in question is practical for GTOM in 3D. / Relativt precisa eyetrackers med ett lågt pris blir allt mer tillgängliga för den vanlige användaren. Således öppnas nya möjligheter upp för att använda ögonspårning i interaktiva realtidsapplikationer. När man förlänger ögats blickpunkt in till virtuella realtidsrenderade 3D-miljöer, används ofta metoder för att kunna korrelera blicken till ett 3D-objekt. Dessa metoder kan benämnas som så kallade gaze-to-object mapping-metoder (GTOM). En GTOM-metod, regionsbundna Interactors, som tillhandahålls av eyetracking-företaget Tobii AB, kompletterades under detta examensarbete med en specifik G-buffer för att kunna extrahera exakta former av enskilda objekt på skärmen. Denna metod jämfördes med två metoder som var baserade på strålföljning i en kontrollerad testmiljö i 3D som bestod av tre typer av olika scenarier med varierande förhållanden. Information samlades in från en användarstudie med 31 deltagare för att kunna bedöma om Interactors-metoden är lämplig för GTOM i virtuella 3D-miljöer. Den utökade metoden uppvisade en del lovande resultat i förhållande till de andra två metoderna, men ytterligare forskning måste utföras för att kunna fastställa om huruvida den aktuella metoden är praktisk för GTOM i 3D.

Thesis (MSc)--Stellenbosch University, 2011. / ENGLISH ABSTRACT: 3D virtual environments can be used as therapy tools in patients with autism spectrum disorders (ASDs); however, the development of such tools is time-consuming. A 3D virtual environment development platform for such tools has been developed specifically for the South African context, because of the language and culture sensitivity of these therapy tools. The 3D virtual environment development platform has a Lua scripting interface for specifying logic in the virtual environments. Lua is a textual programming language, and presents a challenge to ASDs therapists’ ability to create therapy tools without engaging an expert programmer. The aim of this research was to investigate the design and implementation of a visual programming environment to support non-expert programmers in scripting within the 3D virtual environment development platform. Various visual program representation techniques, reported in the literature, were examined to determine their appropriateness for adoption in our design. A visual programming language based on the “building-block” approach was considered the most suitable. The research resulted in the development of a visual script editor (VSE), based on an open source framework called the OpenBlocks library. The VSE successfully alleviated the syntax burden that textual programming languages place on non-expert programmers. The fitness of purpose of our VSE was exemplified in a sample 3D virtual environment that was scripted using the VSE. Despite the success, we argue that the applicability of the “building-block” approach is limited to domain-specific programming languages due to the absence of visual expressions for defining user-defined types, and for specifying hierarchy. / AFRIKAANSE OPSOMMING: Geen opsomming

Made available in DSpace on 2015-03-05T13:58:27Z (GMT). No. of bitstreams: 0 Previous issue date: 10 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O objetivo desta dissertação é propor, testar e avaliar o uso de técnicas de Aprendizado de Máquina (ML) na configuração automática do controle do caminhar de robôs com pernas. Para que este objetivo fosse atingido, um extensa pesquisa de técnicas do estado da arte foi realizada e descrita neste trabalho. Esta pesquisa permitiu a elaboração do modelo proposto, chamado de LegGen, que foi implementado em um protótipo. O protótipo modelo em questão permite a utilização de vários tipos de robôs, compostos de quatro, seis ou mais patas, e além disto permite a evolução da morfologia dos robôs. Utilizando o protótipo, é possível a realização de experimentos com robôs autônomos dotados de pernas, em um ambiente virtual tridimensional realístico, através de simulações baseadas em física. Foi utilizada a biblioteca ODE (Open Dynamics Engine) para a simulação de corpos rígidos e articulações, permitindo assim simular forças agindo nas articulações (atuadores), gravidade e colisões, entre outras propriedades físicas dos / The main goal of this dissertation is to propose, to test and to evaluate the use of Machine Learning (ML) techniques in the automatic con_guration of the gait control in legged robots. In order to achieve this goal, an extensive research about state-of-the-art techniques was accomplished and they are described in this work. This research allowed the development of the proposed model, called LegGen, which was implemented in a prototype. The proposed model allows the use of several different robot models with four, six or more paws. Besides that, the prototype allows also to study the robot's morphology evolution. The implemented prototype allows to accomplish experiments with autonomous legged robots, in a realistic three-dimensional virtual environment, through physics based simulations. The ODE (Open Dynamics Engine) software library was used in the physical simulation of rigid bodies and articulations, allowing to simulate forces acting in the articulations (actuators), gravity and collisions, among other