En quelques années, les domaines des jeux vidéo et des systèmes multi-agents (SMA) ont pris de plus en plus de places dans la vie de chacun. Malgré des similitudes assez fortes (présences d’entité dans les vidéo pouvant être assimilées à des agents), les SMA ne sont presque jamais utilisés dans le développement de jeux. Ce mémoire présente Shine Agent Modeling Platform (SAMP), une plateforme visant intégrer le paradigme multi-agents au sein du développement de jeux vidéo. Cette fusion permet l’utilisation de la puissance des multi-agents au sein des jeux vidéo.SAMP propose une approche au niveau des interactions permettant de réduire le coût de traitement de ces interactions en optimisant le nombre de recherches effectuées dans l’environnement.En plus d’intégrer le paradigme multi-agents, SAMP vise à être accessible à un maximum d’utilisateurs en proposant une interface de modélisation entièrement graphique. Un système d’importation de modèles comportementaux permet de créer deuxniveaux de modélisation : un niveau proche de la logique développement informatique et un niveau proche de la logique métier de l’utilisateur.SAMP est intégré à un moteur de jeux vidéo, Shine Engine, permettant de générer les environnements graphiques dans lesquels les agents évolueront. / In recent years, video games domains and multiagents systems (MAS) domains took more and more place into the life of many pepole. Despite of strong similarities (video games entities wich can be assimilated to agents), MAS are very rarely used during the development of video games. This submission presents the Shine Agent ModelingPlatform (SAMP), a framework trying to integrate the multi-agents paradigm within the development of video games. The purpose is to integrate the efficiency of the MAS within the video games.SAMP provides an approach to enhance the interactions between agents. This approach reduces the number of searches within the environment. In addition to integrate the multi-agents paradigm within the video games, SAMP aims to be userfriendly by proposing a full graphical interface to MAS. An import/export system of these modelsallows users to create two modeling levels: one close to the computer sciences logic and the second close the business logic of the user.SAMP is integrated in a video games engine: Shine Engine. This integration allows to generate the graphic environment in which agents will live.

<div>Cognitive agent architectures embed social learning algorithms and normative frameworks for adopting others’ influenced goals. However, there exists inefficiency in providing continuous, situational decision-making to emerge social, altruistic norms. The thesis reconstructs social-ecological learning mechanisms to functionally and efficiently internalize situational cooperation. By orienting agents to be self-aware of their three-dimensional vectors, i.e., physical, emotional and intellectual in graphical representations, this thesis hypothesizes the parsimonious, action-predictive four emotions that not only link perceptions, action, and cognition by events but also the emotional continuity functional to social-ecological rationality of agents in continuum. Twelve Meridian system is employed to conceptualize the equalizing, complementary, heuristic orientation (ECHO) model. ECHO simulates “naturalistic” cooperation to model embodied, social-ecological orientations by self-organizing emotions to emerge functional social network formations. ECHO delineates the soma links to perceptions, namely Twelve Meridian channels as “direct pipes” that initiates and conduct emotions and consciousness of three dimensional agenthood: physical, emotional, and intellectual desires. ECHO reconstructs emotions as entities to induce systemic, self-organized rule of delegation by integrating agents’ percepts and actuations. By modeling constitutional emotions and consciousness of eight entities, emotions within entities as “individualized emotional processors,” are constructing and integrating purposeful social, altruistic events for the efficacy of situated agents.<br></div>

Artificial intelligence (AI) agents will need to interact with both other AI agents and humans. One way to enable effective interaction is to create models of associates to help to predict the modeled agents' actions, plans, and intentions. If AI agents are able to predict what other agents in their environment will be doing in the future and can understand the intentions of these other agents, the AI agents can use these predictions in their planning, decision-making and assessing their own potential. Prior work [13, 14] introduced the S# algorithm, which is designed as a robust algorithm for many two-player repeated games (RGs) to enable cooperation among players. Because S# generates actions, has (internal) experts that seek to accomplish an internal intent, and associates plans with each expert, it is a useful algorithm for exploring intent, plan, and action in RGs. This thesis presents a graphical Bayesian model for predicting actions, plans, and intents of an S# agent. The same model is also used to predict human action. The actions, plans and intentions associated with each S# expert are (a) identified from the literature and (b) grouped by expert type. The Bayesian model then uses its transition probabilities to predict the action and expert type from observing human or S# play. Two techniques were explored for translating probability distributions into specific predictions: Maximum A Posteriori (MAP) and Aggregation approach. The Bayesian model was evaluated for three RGs (Prisoners Dilemma, Chicken and Alternator) as follows. Prediction accuracy of the model was compared to predictions from machine learning models (J48, Multi layer perceptron and Random Forest) as well as from the fixed strategies presented in [20]. Prediction accuracy was obtained by comparing the model's predictions against the actual player's actions. Accuracy for plan and intent prediction was measured by comparing predictions to the actual plans and intents followed by the S# agent. Since the plans and the intents of human players were not recorded in the dataset, this thesis does not measure the accuracy of the Bayesian model against actual human plans and intents. Results show that the Bayesian model effectively models the actions, plans, and intents of the S# algorithm across the various games. Additionally, the Bayesian model outperforms other methods for predicting human actions. When the games do not allow players to communicate using so-called cheaptalk, the MAP-based predictions are significantly better than Aggregation-based predictions. There is no significant difference in the performance of MAP-based and Aggregation-based predictions for modeling human behavior when cheaptalk is allowed, except in the game of Chicken.

Animated agent technology has been rapidly developed to provide ubiquitously psychological and functional benefits for fulfilling communicative goals. However, the character motions of most character-centered models based on pre-stored movement, finite state machine and scripted conditional logic are generally restrictive. The major drawback lies in the lack of maturity of integrating the elements between personality, emotion and behaviour. To bridge the gap between cognitive and behavioural elements, we examine the connections between human personality, emotion, movement and cartoon modeling for the agent design. Human personality and emotional behaviour are the essences in the recognition of a believable synthetic character. Personality and emotion come from the storylines and result in characters’ motions. Cartoon animations successfully engage the audience and create emotional connections with the spectators. However, even a sophisticated animator often faces some difficulties while performing a very laborious task to simulate an emotion- and personality-rich character. This thesis focuses on exploring effective techniques to extract personality and emotion features for a high-level control of character movements. A hierarchical fuzzy rule-based system was constructed, in which personality and emotion were mapped into the body’s movement zones of a character. This facilitates agent designers to control the personality and emotion of a dynamic synthetic character. The system was then applied to a Narrative Intelligent system and extended to an Ambient Intelligent environment. An innovative storyboard-structured storytelling method was devised by using story scripts and action descriptions in a form similar to the content description of storyboards to predict specific personality and emotion. As software or device agents evolve into the Ambient Intelligence, new concepts for effective agent presentations and delegating control are necessary to minimise the human’s tasks and interventions in the complex and dynamic environment. A novel customizable personalised agent framework was developed by utilising the spirit of cartoon animation to match each user’s profile in the form of a cartoon reciprocal agent. As a result, users could explicitly modify personality and emotion values to change the psychology traits of the agent, which would affect their appearance and behaviour through body posture expression. An evaluation of the system was conducted to verify the effectiveness and the applicability in both Narrative and Ambient intelligent agent frameworks. The significance of this research is that applying higher cognitive factors to animated characters can lead to a better animation design tool and reduce strenuous animation production efforts in agent designs. It will also enable animated characters to embody more adaptive, flexible and stylised performance.

This thesis describes the paradigm of complex systems and discusses possibilities of their modeling and simulations. The work shows the suitability of using multi-agent modeling for creating abstraction of social environment, that is one of the major complex systems. Thesis content includes an analysis of tools that are available for creating multi-agent simulators. The core of this thesis are processed research documents and a demonstrative model of social phenomenon known as procrastination.

Most story-based games today have manually-scripted non-player characters (NPCs) and the scripts are usually simple and repetitive since it is time-consuming for game developers to script each character individually. ScriptEase, a publicly-available author-oriented developer tool, attempts to solve this problem by generating script code from high-level design patterns, for BioWare Corp.'s role-playing game Neverwinter Nights. The ALeRT algorithm uses reinforcement learning (RL) to automatically generate NPC behaviours that change over time as the NPCs learn from the successes or failures of their own actions. This thesis aims to provide a new learning mechanism to game agents so they are capable of adapting to new behaviours based on the actions of other agents. The new on-line RL algorithm, ALeRT-AM, which includes an agent-modeling mechanism, is applied in a series of combat experiments in Neverwinter Nights and integrated into ScriptEase to produce adaptive behaviour patterns for NPCs.

No description available.

Актуальность темы магистерской диссертации заключается в ее тесной связи с глобальной пандемией нового коронавируса, при этом особое внимание уделяется распространению эпидемии и борьбе с ней. Целью исследования является предоставление научной и научно обоснованной поддержки путем разработки моделей и симуляций политики профилактики эпидемий в сообществе. Основная цель диссертационной работы – оценить факторы, влияющие на эффективность стратегий профилактики и контроля, а также раскрыть ключевые факторы и механизмы передачи эпидемии. Посредством симуляционных экспериментов, анализа и сравнения результатов создается исчерпывающая информация, которая поможет лицам, принимающим решения, формулировать и осуществлять более эффективную политику профилактики эпидемий на уровне сообщества. Целью данного исследования является изучение влияния мобильности населения и планировки жилого массива на передачу заболеваний и эффективность стратегий профилактики эпидемий. Предметом исследования является разработка системы моделирования и симуляции политики предотвращения эпидемий на уровне сообщества с использованием модели SIR и сети «малого мира» в жилом сообществе с численностью населения 500 человек. Научная новизна данного исследования заключается в сочетании классической модели SIR с сетевой моделью маленького мира, а также в использовании агентной модели и программного обеспечения NetLogo для моделирования. Этот инновационный подход учитывает взаимодействие и связи между людьми в сообществе, позволяя более точно моделировать распространение болезней и оценивать эффекты различных стратегий профилактики эпидемий. Практическая значимость исследования заключается в обеспечении научной основы и руководства для лиц, принимающих решения. Путем проведения симуляционных экспериментов и анализа результатов исследование оптимизирует разработку и реализацию политики профилактики эпидемий на уровне сообщества, эффективно контролируя распространение заболеваний, защищая здоровье населения и решая проблемы, связанные с инфекционными заболеваниями. / The relevance of the master's thesis topic lies in its close connection to the global novel coronavirus pandemic, specifically focusing on the spread and control of the epidemic. The research aims to provide scientific and evidence-based support by developing community epidemic prevention policy models and simulations. The main goal of the thesis is to evaluate the factors influencing the effectiveness of prevention and control strategies and uncover the key factors and mechanisms of epidemic transmission. Through simulation experiments, analysis, and comparison of results, comprehensive information is generated to assist decision makers in formulating and implementing more effective community epidemic prevention policies. The objective of this study is to examine the influence of population mobility and the layout of a residential community on disease transmission and the effectiveness of epidemic prevention strategies. The subject of research focuses on developing a modeling and simulation framework for community epidemic prevention policies using the SIR model and small-world network in a residential community with a population size of 500 individuals. The scientific novelty of this study lies in the combination of the classic SIR model with the small-world network model, along with the introduction of the agent model and NetLogo software for simulation. This innovative approach considers the interactions and connections between individuals in a community, enabling a more accurate modeling of disease spread and evaluation of the effects of different epidemic prevention policies. The practical significance of the research lies in its provision of scientific basis and guidance to decision makers. By conducting simulation experiments and analyzing the results, the study optimizes the formulation and implementation of community epidemic prevention policies, effectively controlling the spread of diseases, protecting public health, and addressing the challenges posed by infectious diseases.

Made available in DSpace on 2016-08-17T14:52:59Z (GMT). No. of bitstreams: 1 Marcos Barros e Silva.pdf: 1611676 bytes, checksum: 2ac174c844a81425dc12f950d38f1b39 (MD5) Previous issue date: 2005-06-28 / This dissertation is part of TMS project whose objective is the development of a Intelligent Traffic Administration System . TMS has three fundamental functions: (1) Manage the necessary resources to the traffic control; (2) aid in the training of traffic employees; and (3) Accompany the changes of the urban traffic for ends of decision. TMS was conceived to assist the cities that doesn't have systems of traffic control and the others witch has the whole mounted structure with interlinked traffic lights. We show the conception of a system that seeks to assist the needs of traffic administration organs and, at the same time, allow its users to aid in the constant evolution of this management. We will specifically treat an architecture based on agents for our system, the technologies which the system uses and interactions among agents of traffic control. The prototype implemented in this work it allows the automation of the mechanisms of Traffic Administration operation - time of green and red of the traffic lights - turning the changes of semaphored plans the more dynamics as possible. / Esta dissertação é parte do Projeto TMS cujo objetivo é o desenvolvimento de um Sistema inteligente de Gerenciamento de Trânsito. O TMS tem três funções fundamentais : (1) Gerenciar os recursos necessários ao controle do tráfego ; (2) Ajudar no treinamento dos funcionários das companhias de trânsito ; e (3) Acompanhar as mudanças do tráfego urbano para fins de tomada de decisão. O TMS foi concebido para atender desde as cidades que não tem sistemas de controle de tráfego até aquelas que tem todo a estrutura montada com semáforos interligados, etc. Apresenta-se aqui a concepção do sistema que visa atender às necessidades dos órgãos de gerenciamento de trânsito e que, ao mesmo tempo, permita que os seus usuários pudessem auxiliar na constante evolução do mesmo. Trataremos especificamente da arquitetura baseada em agentes proposta para o sistema, as tecnologias que o fundamentam e apreendemos as interações entre agentes de controle de trânsito. O protótipo implementado neste trabalho permite a automação dos mecanismos de funcionamento de Gerenciamento de Trânsito tempo de verde e vermelho dos semáforos - tornando as mudanças de planos semafóricos as mais dinâmicas possíveis.