No description available.

Navigation mesh is a form of representation of a virtual environment, which allows effective navigation of intelligent virtual agents. This thesis deals with the development of a navigation mesh for the intelligent virtual agents in the environment of the computer game Unreal Tournament 2004. Before the development of the navigation mesh, the navigation graph was the only available representation of the environment that the agents could use for navigation. The navigation graph however provides only limited information about the environment. With the use of the navigation mesh the agents are able to navigate in the environment faster and over shorter paths. Creation of the navigation mesh from the environment's geometry extracted from Unreal Tournament 2004 map files is a part of the thesis. Finally, a new navigation module for Pogamut 3 - a platform that allows creation and controlling of intelligent virtual agents, is implemented. The new navigation module consists of the navigation mesh and a set of navigation algorithms Powered by TCPDF (www.tcpdf.org)

Title: Level of Detail AI: a prototype of a large virtual world Author: David Skupien Department: Department of Software and Computer Science Education Supervisor: Mgr. Cyril Brom, Ph.D. Abstract: The Level Of Detail technique (LOD) is an established means of optimization in computer graphics. It uses the limitations of human perception to spare system resources. Progressively, this approach has found its use in artificial intelligence. This adaptation, commonly known as LOD AI, is used to optimize autonomous agent control in large virval worlds. This thesis consists of two parts. The first part deals with standing LOD AI uses, mostly focusing on the IVE project. The second part directly uses the IVE project to implement a LOD AI game using the LOD based reactive planning algorithm called S-GHRP over the ISMA representation. The above game is then tested on respondents to verify the practical usability of S-GHRP/ISMA. In its conclusion, this thesis briefly ponders the possibilities of further development of techniques used in the IVE project, based on previous findings. Keywords: artificial intelligence, LOD AI, virtual agent, large virtual world, IVE

Submitted by JOSIANE SANTOS DE OLIVEIRA (josianeso) on 2019-01-22T11:23:51Z No. of bitstreams: 1 Renato Laurentino_.pdf: 2042175 bytes, checksum: 337314ffc10cc5bf0976577e8911e870 (MD5) / Made available in DSpace on 2019-01-22T11:23:52Z (GMT). No. of bitstreams: 1 Renato Laurentino_.pdf: 2042175 bytes, checksum: 337314ffc10cc5bf0976577e8911e870 (MD5) Previous issue date: 2018-08-29 / UNIDAVI - Centro Universitário para o Desenvolvimento do Alto Vale do Itajaí / Evolução tecnológica, uso de "robôs" nos atendimentos, consumidores mais exigentes, ativos e com comportamentos mais complexos e difíceis de serem previstos, enxergar o consumidor além de um mero receptor passivo, são alguns dos desafios a serem enfrentados pelas organizações. Compreender os motivadores comportamentais dos consumidores é fator imprescindível para a competitividade de um negócio. Este trabalho buscou analisar a intenção de uso da tecnologia de atendimento autônomo e a disposição do consumidor em cocriar quando no uso dos serviços prestados por um Agente Virtual de Atendimento (AVA), tendo por mediadores do processo, no modelo proposto, a percepção de utilidade, facilidade de uso e atitude. Como antecedentes, no modelo construído com base na revisão da literatura, seis foram propostos (Normas Subjetivas, Necessidade de Interação, Experiência, Auto-eficácia, Controle e Ansiedade). O método utilizado neste trabalho é o quantitativo com uma abordagem conclusiva descritiva que utilizou um levantamento com 943 respondentes de todo o Brasil, com idades entre 15 e 65 anos, a partir de uma survey com escala intervalar de sete pontos. A análise dos resultados foi feita com base em estatísticas multivariadas, utilizando-se como método de análise a Modelagem de Equações Estruturais no intuito de testar o modelo, investigar e analisar os elementos propostos e suas relações. O trabalho apresentou 18 hipóteses das quais 12 apresentaram-se significantes e 6 não apresentaram níveis de significância adequados. Como resultado mais relevante encontrou-se um forte caminho que propõe uma influência significante e positiva da facilidade de uso sobre a utilidade (sig<0,001 e β = 0,543), utilidade influenciando a atitude (sig<0,001 e β = 0,745), atitude impactando positivamente a intenção de uso (sig<0,001 e β = 0,917) e intenção de uso influenciando positivamente a disposição para cocriar (sig<0,001 e β = 0,656). Esta pesquisa demonstrou que o consumidor percebe-se inserido no processo de cocriação e encontra valor nesta relação, mesmo quando no uso de serviços prestados por Agentes Virtuais de Atendimento. Demonstrou também que um Agente Virtual de Atendimento pode ser uma boa solução na prestação de serviços uma vez que seja útil e eficaz do ponto de vista dos consumidores. / Technological evolution, the use of "robots" in the services, more demanding consumers, more active and with more complex behaviors and difficult to be foreseen, to see the consumer as well as a mere passive receiver are some of the challenges to be faced by the organizations. Understanding the behavioral motivators of consumers is an essential factor for the competitiveness of a business. This work sought to analyze the intention to use the technology of autonomous service and the willingness of the consumer to co-create when using the services provided by a Virtual Attendance Agent (AVA), having as mediators of the process, in the proposed model, the perception of utility , ease of use and attitude. As background, in the model built on the literature review, six were proposed (Subjective Norms, Need for Interaction, Experience, Self-efficacy, Control and Anxiety). The method used in this work is the quantitative one with a conclusive descriptive approach that used a survey with 943 respondents from all over Brazil, aged between 15 and 65 years, from a survey with an interval of seven points. The analysis of the results was made based on multivariate statistics, using as a method of analysis the Modeling of Structural Equations in order to test the model, investigate and analyze the proposed elements and their relationships. The study presented 18 hypotheses of which 12 were significant and 6 did not present adequate levels of significance. As a more relevant result, we found a strong path that proposes a significant and positive influence of ease of use on utility (sig <0.001 and β = 0.543), utility influencing attitude (sig <0.001 and β = 0.745), attitude impacting positively the intention to use (sig <0.001 and β = 0.917) and intention of use positively influencing the willingness to co-create (sig <0.001 and β = 0.656). This research demonstrated that the consumer perceives himself inserted in the process of co-creation and finds value in this relation, even when in the use of services provided by Virtual Agents of Attendance. It has also demonstrated that a Virtual Attendance Agent can be a good service delivery solution as it is useful and effective from the point of view of consumers.

L'identification des visages est primordiale lors de nos interactions sociales. Ainsi, notre comportement change suite à l'identification de la personne avec laquelle nous interagissons. De plus, les travaux en psychologie et en neurosciences ont observé que le traitement cognitif face à un visage familier diffère de celui que nous avons face à un visage inconnu.D'une autre part, les dernières techniques de rendu 3D et les dernières avancées des scans 3D ont permis la création de visages virtuels photo-réalistes modélisant des personnes réelles existantes. La tendance actuelle pour modéliser des humains virtuels est de se baser sur des techniques d'acquisition de données réelles (issues de scans et de sessions de capture de mouvement). Par conséquent, les recherches et applications en humains virtuels ont connu un intérêt croissant pour ces clones virtuels (des agents ayant un aspect familier ou du moins reconnaissable). Les clones virtuels sont donc de plus en plus répandus dans des interfaces homme-machine et dans l'industrie audio-visuelle.L'étude de la perception et de l'interaction avec des clones virtuels est donc devenue nécessaire pour appréhender la conception et l'évaluation de cette technologie. En effet, très peu d'études se sont penchées sur l'évaluation de la fidélité de ces clones virtuels. L'objectif de cette thèse consiste à explorer cet axe de recherche en examinant le processus de perception de la fidélité d'un visage virtuel, clone d'une personne réelle (que l'on connait ou non).Nos travaux répondent à plusieurs questions de recherche: Quels sont les éléments qui nous permettent d'évaluer la ressemblance du clone virtuel avec son référent? Parmi les multiples possibilités de techniques de rendu, d'animation et d'acquisition de données qu'offre l'informatique graphique, quelle est la meilleure combinaison pour assurer le plus haut degré de fidélité perçue ? L'apparence visuelle n'est cependant qu'une des composantes qui interviennent dans la reconnaissance de personnes familières. Les autres composantes comprennent ainsi l'expressivité mais aussi le traitement des connaissances que nous avons sur cette personne (par exemple sa manière particulière d'évaluer une situation émotionnelle et de l'exprimer via son visage).Nos contributions apportent des éléments de réponse à ces questions à plusieurs niveaux. Nous avons défini un cadre conceptuel identifiant les principaux concepts pertinents pour l'étude de la fidélité d'un visage virtuel. Nous avons aussi étudié l'aspect visuel de la fidélité à travers l'exploration de différentes techniques de rendu. Nous avons étudié dans une autre étape l'impact de la familiarité dans le jugement de la fidélité. Finalement, nous avons proposé un modèle informatique individuel basé sur une approche cognitive des émotions qui permet de guider l'animation expressive du clone virtuel.Ces travaux de thèse ouvrent des perspectives pour la conception et l'amélioration de clones virtuels, mais aussi plus généralement des interfaces homme-machine basées sur des agents expressifs. / Face identification plays a crucial role in our daily social interactions. Indeed, our behavior changes according to the identification of the person with whom we interact. Moreover, several studies in Psychology and Neurosciences have observed that our cognitive processing of familiar faces is different from the cognitive processing of unfamiliar faces.Creating photorealistic an animated human-like face of a real person is now possible thanks to recent advances in Computer Graphics and 3D scan systems. Recent rendering techniques are challenging our ability to distinguish between computer generated faces and real human faces. Besides, the current trend to model virtual humans is to involve real data collected using scans and motion capture systems. Research and applications in virtual humans have experienced a growing interest in so-called virtual clones (agents with a familiar aspect or at least recognizable). Virtual clones are therefore increasingly used in human-machine interfaces and in the audiovisual industry. Studies about the perception and interaction with virtual clones are therefore required to better understand how we should design and evaluate this kind of technology. Indeed, very few studies have tried to evaluate virtual clones' fidelity with respect to the original human (hereafter called “the referent”). The main goal of this thesis is to explore this line of research. Our work rises several research questions: What are the features of the virtual clone that enable us to evaluate the resemblance between a virtual clone and its referent? Among several possibilities of rendering, animation and data acquisition techniques offered by Computer Graphics, what is the best combination of techniques to ensure the highest level of perceived fidelity?However, visual appearance is not the only component that is involved in recognizing familiar people. The other components include facial expressiveness but also the possible knowledge that we have about the referent (e.g. his particular way of assessing an emotional situation and expressing it through his face).Our contributions provide answers to these questions at several levels. We define a conceptual framework identifying the key concepts which are relevant for the study of the fidelity of a virtual face. We explore different rendering techniques. We describe an experimental study about the impact of familiarity in the judgment of fidelity. Finally, we propose a preliminary individual computational model based on a cognitive approach of emotions that could drive the animation of the virtual clone.This work opens avenues for the design and improvement of virtual clones, and more generally for the human-machine interfaces based on expressive virtual agents.

The purpose of this research study was to explore the impact of virtual character design on user emotional experience and user behavior in a simulated environment. With simulation training increasing in popularity as a tool for teaching social skills, it is essential that social interactions in virtual environments provide authentic opportunities for practice (Swartout et al., 2006). This study used Interactive Performance Theory (Wirth, 2012) to examine the effect of designing a virtual buddy character with ineffective traits instead of effective or expert traits. The sample population for this study (n = 145) consisted of first year university students enrolled in courses in the fall of 2013 at the University of Central Florida. Data on participant emotional experience and behavior were collected through questionnaires, researcher observations, and physiological signal recording that included participant heart rate and galvanic skin response. Data were analyzed using multivariate analysis of variances (MANOVA), Kruskal-Wallis one-way analysis of variance, and qualitative thematic coding of participant verbal behavior and written responses. Results of the analysis revealed that participants who interacted with an ineffective virtual buddy character had statistically significant higher averages of verbal statements to the antagonist in the simulated environment and statistically significant lower perceptions of antagonist amiability than participants who interacted with an effective virtual buddy. Additionally, participants who interacted with a virtual buddy of the opposite gender gave statistically significant higher ecological validity scores to the simulated environment than participants who interacted with a virtual buddy of the same gender. Qualitative analysis also revealed that participants tended to describe the female buddy character with more ineffective traits than the male buddy character even though effective and ineffective design conditions were equally divided for both groups. Further research should be conducted on the effect of virtual buddy character design in different types of simulation environments and with different target audiences.

<p dir="ltr">With the recent virtualization of our everyday lives and the development of intelligent AI technology, realistic virtual agents are becoming a useful tool for research, education, and entertainment. With virtual agents customized content can be created and individualized for specific users. However, virtual agents are only effective tools if they can form a connection with the individual. To form a connection the virtual agent must be believable.</p><p dir="ltr">The literature review evaluates topics of virtual agent believability, appeal, and naturalness and how they relate to asymmetry in facial expression animation. The literature suggests that asymmetries can affect the perception of virtual agents. Additionally, it suggests that emulating human behavior is beneficial to increasing perception of believability, appeal, and naturalness.</p><p dir="ltr">In this study we evaluated the effects of facial expression asymmetry on the believability, appeal, and naturalness of virtual agents. To do this we ran an online perception study with students at Purdue University. We found, in brief, that facial expression asymmetries do have significant effect on the believability, appeal, and naturalness of a virtual agent compared to animations that do not include facial expression asymmetries.</p>

A framework for understanding the social responses to virtual humans suggests that human-like characteristics (e.g., facial expressions, voice, expression of emotion) act as cues that lead a person to place the agent into the category "human" and thus, elicit social responses. Given this framework, this research was designed to answer two outstanding questions that had been raised in the research community (Moon&Nass, 2000): 1) If a virtual human has more human-like characteristics, will it elicit stronger social responses from people? 2) How do the human-like characteristics interact in terms of the strength of social responses? Two social psychological (social facilitation and politeness norm) experiments were conducted to answer these questions. The first experiment investigated whether virtual humans can evoke a social facilitation response and how strong that response is when participants are given different cognitive tasks (e.g., anagrams, mazes, modular arithmetic) that vary in difficulty. They did the tasks alone, in the company of another person, or in the company of a virtual human that varied in terms of features. The second experiment investigated whether people apply politeness norms to virtual humans. Participants were tutored and quizzed either by a virtual human tutor that varied in terms of features or a human tutor. Participants then evaluated the tutor's performance either directly by the tutor or indirectly via a paper and pencil questionnaire. Results indicate that virtual humans can produce social facilitation not only with facial appearance but also with voice recordings. In addition, performance in the presence of voice synced facial appearance seems to elicit stronger social facilitation (i.e., no statistical difference compared to performance in the human presence condition) than in the presence of voice only or face only. Similar findings were observed with the politeness norm experiment. Participants who evaluated their tutor directly reported the tutor's performance more favorably than participants who evaluated their tutor indirectly. In addition, this valence toward the voice synced facial appearance had no statistical difference compared to the valence toward the human tutor condition. The results suggest that designers of virtual humans should be mindful about the social nature of virtual humans.

This case study investigates a chatbot and its responsiveness to user expression, utility and the users' experience. A user study was performed with both a questionnaire and user interviews. A program to automatically categorize user input messages was made in order to analyze the correlations and reasons for failed predictions and misunderstandings. The current utility is not reaching the return on investment, although expansions are planned and the launch is underway. The strongest and most meaningful correlations with unsuccessful messages were with messages that were out of the chatbot's scope, inside of the scope but unsolved, messages that lacked training and test data or that contained at least one unknown word. The user rated the chatbot's ability to understand them slightly above the middle of the scale, even though the rate of unsuccessful messages was high since the chatbot is still under development. Misspellings, vernacular and organization specific words did not have a particularly strong correlation to unsuccessful messages. The opinions on humanlike interaction were split both in regard to its current likeness and what would be desirable in the future. However, most users seemed to want the conversation to feel natural as if from a human and to receive nuanced answers.

Digitala kommunikationstjänster som chattbotar och virtuella assistenter blir alltmer förekommande och populärt för företag och organisationer att använda sig av. Chattboten och den virtuella assistenten är besläktade datorsystem uppbyggda med bakomliggande teknik som syftar till att efterlikna mänskliga samtal på ett så realistiskt sätt som möjligt. Företag och organisationer använder dessa digitala kommunikationsverktyg både för interna- och externa bruk som ett kundtjänstalternativ. Vid användning av dessa digitala kommunikationstjänster tillkommer en rad utmaningar. Denna studie ämnar ta reda på de främst förekommande för- och nackdelarna. Studien har genomförts i form av intervjuer med sex svenska storföretag som implementerat chattboten eller den virtuella assistenten på sin webbplats och/eller använder den som ett hjälpmedel internt. En av studiens intervjuade kandidater är utvecklare av virtuell assistans och använder inte verktyget själva. Studiens insamlade resultat från intervjuerna visar att de främst förekommande fördelarna med en chattbot eller virtuell assistent är 1) Tillgängligheten dygnet runt, 2) Den smidiga övergången från chatt till mänsklig handläggare, och 3) Hantering av monotona enkla arbetsuppgifter. De främst förekommande nackdelarna är 1) Systemet tränas med felaktiga data, 2) Olika information når kunden från chattboten och en mänsklig handläggare, och 3) För höga förväntningar på systemet från början. Studiens sammanfattade slutsats är att chattboten och den virtuella assistenten kräver kontinuerligt bakomliggande arbete för att upprätthålla en så god användarupplevelse som möjligt. / Digital communication services such as chatbots and virtual assistants are becoming more common and popular for companies and organizations to use. The chatbot and the virtual assistant are related computer systems built with underlying technology that aims to mimic human conversations in a way as realistic as possible. Companies and organizations use these digital communication tools for both internal and external use as a customer service alternative. When using these digital communication services, several challenges arise. This study intends to find out the main advantages and disadvantages. The study was conducted in the form of interviews with six Swedish companies that have implemented the chatbot or the virtual assistant on their website and/or use it as assistance internally. One of the study's interviewed candidates is a developer of virtual assistance and does not use it themselves within the company. The study's collected results from the interviews show that the main advantages of a chatbot or virtual assistant are 1) Accessibility around the clock, 2) The smooth transition from chat to human administrator, and 3) Handling of monotonous simple tasks. The main disadvantages are 1) The system is trained with incorrect data, 2) Different information reaches the customer from the chatbot and a human administrator, and 3) Too high expectations of the system from the beginning. The study's summary conclusion is that the chatbot and the virtual assistant require continuous underlying work to maintain as good a user experience as possible.