Game Environment Design Creator using Artificial Intelligence Procedural Generation

Detalhes bibliográficos
Autor(a) principal: Auxtero, André Linard Santos
Data de Publicação: 2023
Tipo de documento: Dissertação
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: http://hdl.handle.net/10362/164247
Resumo: Automatic generation is a prevalent technique in video games, applied to various elements like map design, character creation and level complexity. The current emergence of Artificial Intelligence (AI), such as Chat GPT and Large Language Models (LLM), has revolutionized problem solving in multiple domains, leading to the emerging of courses dedicated to mastering prompt based AI. This thesis investigates into the application of Artificial Intelligence, specifically LLM, to implement automatic environment generation within a system for interactive story creation in a video game format. The primary objective is to develop an intuitive system that allows users to effortlessly, create and experience different stories without requiring prior gaming or programming knowledge. This is achieved by simplifying complex game and story creation tasks, particularly, environment design, using an algorithm that automates the initial process while still allowing the user customization. This thesis explores multiple methods of Automatic Generation, drawing inspiration from Procedural Generation algorithms and Artificial Intelligence. Additionally, investigates multiple gaming related concepts, including Realistic and Semi- Realistic/Unrealistic design and the exploration of OpenWorld and Modular Architecture game formats. The target audience are the individuals interested in creating andpresenting their narratives in a video game framework. While the primary focus is on adventure games due to their tendency for structured storytelling, the system’s versatility also allows for the creation of narratives including different genres, including educational narratives and re imaginings of well-known tales. The final prototype is a comprehensive system that uses LLM to automatically generate customizable environments. Stories can be experienced in a 3D point and click, topdown view, and the system is able to create diverse narratives, including a pizza restaurant story line and a re-creation of a segment from the game "The Secret of Monkey Island". This thesis exemplifies the potential of AI driven automatic generation, more specifically with AI-powered generative language models to facilitate accessible and imaginative storytelling in the realm of video games.
id RCAP_b203ce3fdcac7681a62eed737f7e439d
oai_identifier_str oai:run.unl.pt:10362/164247
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling Game Environment Design Creator using Artificial Intelligence Procedural GenerationAutomatic generationArtificial intelligenceAI-powered generative language modelsVideo gamesCustomizable EnvironmentsInteractive storytellingDomínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e InformáticaAutomatic generation is a prevalent technique in video games, applied to various elements like map design, character creation and level complexity. The current emergence of Artificial Intelligence (AI), such as Chat GPT and Large Language Models (LLM), has revolutionized problem solving in multiple domains, leading to the emerging of courses dedicated to mastering prompt based AI. This thesis investigates into the application of Artificial Intelligence, specifically LLM, to implement automatic environment generation within a system for interactive story creation in a video game format. The primary objective is to develop an intuitive system that allows users to effortlessly, create and experience different stories without requiring prior gaming or programming knowledge. This is achieved by simplifying complex game and story creation tasks, particularly, environment design, using an algorithm that automates the initial process while still allowing the user customization. This thesis explores multiple methods of Automatic Generation, drawing inspiration from Procedural Generation algorithms and Artificial Intelligence. Additionally, investigates multiple gaming related concepts, including Realistic and Semi- Realistic/Unrealistic design and the exploration of OpenWorld and Modular Architecture game formats. The target audience are the individuals interested in creating andpresenting their narratives in a video game framework. While the primary focus is on adventure games due to their tendency for structured storytelling, the system’s versatility also allows for the creation of narratives including different genres, including educational narratives and re imaginings of well-known tales. The final prototype is a comprehensive system that uses LLM to automatically generate customizable environments. Stories can be experienced in a 3D point and click, topdown view, and the system is able to create diverse narratives, including a pizza restaurant story line and a re-creation of a segment from the game "The Secret of Monkey Island". This thesis exemplifies the potential of AI driven automatic generation, more specifically with AI-powered generative language models to facilitate accessible and imaginative storytelling in the realm of video games.A geração automática é uma técnica prevalente em videojogos, aplicada a vários elementos, como o design de cenários, a criação de personagem e a dificuldade dos níveis. O surgimento da Inteligência Artificial, o Chat GPT ou Large Language Models (LLM), revolucionou a resolução de problemas em diversas áreas. Começam a surgir cursos dedicados a dominar o prompting necessário para utilizar estes modelos. Esta tese explora a aplicação da Inteligência Artificial, especificamente LLM, para implementar a geração automática de cenários dentro de um sistema projetado para contar histórias interativas num formato de videojogo. O principal objetivo é desenvolver um sistema intuitivo que permita aos utilizadores cirar e experienciar estórias diversas sem exigir conhecimentos prévios na área de jogos ou programação. Isto é alcançado através da simplificação de tarefas complexas de criação de jogos e histórias, nomeadamente o design do cenário, por meio de um algoritmo que automatiza esse processo, ao mesmo tempo que permite ser customizável pelo utilizador. A tese explora múltiplos métodos de Geração Automática, inspirando-se em algoritmos de Geração Procedimental e Inteligência Artificial. Além disso, investiga diversos conceitos relacionados com jogos, incluindo abordagens de design Realistas e Semi-Realistas/Irrealistas e a exploração de formatos de jogos em Arquitetura de Mundo Aberto e Modular. O público-alvo é composto por indivíduos interessados em criar e apresentar as suas narrativas num contexto de videojogo. Embora o foco principal seja jogos de aventura devido à sua tendência para contar estórias bem estruturadas, a versatibilidade do sistema permite a criação de estórias de diferentes géneros, incluindo narrativas educativas e reinterpretações de contos conhecidos. O protótipo final é um sistema abrangente que usa LLM para gerar automaticamente cenários, permitindo ajustes pelo utilizador. As estórias podem ser vivenciadas numa perspetiva 3D point and click, em vista topdown e o sistema é capaz de criar narrativas diversas, incluindo uma história num restaurante de pizzas e uma recriação de um segmento do jogo "The Secret of Monkey Island". Esta tese exemplifica o potencial da geração automática impusionada por modelos linguísticos generativos por IA, para facilitar a criação de estórias no domínio dos videojogos.Vieira, JoãoNóbrega, RuiRUNAuxtero, André Linard Santos2024-02-28T15:31:57Z2023-122023-12-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10362/164247enginfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2024-03-11T05:51:35Zoai:run.unl.pt:10362/164247Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-20T04:00:06.734460Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse
dc.title.none.fl_str_mv Game Environment Design Creator using Artificial Intelligence Procedural Generation
title Game Environment Design Creator using Artificial Intelligence Procedural Generation
spellingShingle Game Environment Design Creator using Artificial Intelligence Procedural Generation
Auxtero, André Linard Santos
Automatic generation
Artificial intelligence
AI-powered generative language models
Video games
Customizable Environments
Interactive storytelling
Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática
title_short Game Environment Design Creator using Artificial Intelligence Procedural Generation
title_full Game Environment Design Creator using Artificial Intelligence Procedural Generation
title_fullStr Game Environment Design Creator using Artificial Intelligence Procedural Generation
title_full_unstemmed Game Environment Design Creator using Artificial Intelligence Procedural Generation
title_sort Game Environment Design Creator using Artificial Intelligence Procedural Generation
author Auxtero, André Linard Santos
author_facet Auxtero, André Linard Santos
author_role author
dc.contributor.none.fl_str_mv Vieira, João
Nóbrega, Rui
RUN
dc.contributor.author.fl_str_mv Auxtero, André Linard Santos
dc.subject.por.fl_str_mv Automatic generation
Artificial intelligence
AI-powered generative language models
Video games
Customizable Environments
Interactive storytelling
Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática
topic Automatic generation
Artificial intelligence
AI-powered generative language models
Video games
Customizable Environments
Interactive storytelling
Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática
description Automatic generation is a prevalent technique in video games, applied to various elements like map design, character creation and level complexity. The current emergence of Artificial Intelligence (AI), such as Chat GPT and Large Language Models (LLM), has revolutionized problem solving in multiple domains, leading to the emerging of courses dedicated to mastering prompt based AI. This thesis investigates into the application of Artificial Intelligence, specifically LLM, to implement automatic environment generation within a system for interactive story creation in a video game format. The primary objective is to develop an intuitive system that allows users to effortlessly, create and experience different stories without requiring prior gaming or programming knowledge. This is achieved by simplifying complex game and story creation tasks, particularly, environment design, using an algorithm that automates the initial process while still allowing the user customization. This thesis explores multiple methods of Automatic Generation, drawing inspiration from Procedural Generation algorithms and Artificial Intelligence. Additionally, investigates multiple gaming related concepts, including Realistic and Semi- Realistic/Unrealistic design and the exploration of OpenWorld and Modular Architecture game formats. The target audience are the individuals interested in creating andpresenting their narratives in a video game framework. While the primary focus is on adventure games due to their tendency for structured storytelling, the system’s versatility also allows for the creation of narratives including different genres, including educational narratives and re imaginings of well-known tales. The final prototype is a comprehensive system that uses LLM to automatically generate customizable environments. Stories can be experienced in a 3D point and click, topdown view, and the system is able to create diverse narratives, including a pizza restaurant story line and a re-creation of a segment from the game "The Secret of Monkey Island". This thesis exemplifies the potential of AI driven automatic generation, more specifically with AI-powered generative language models to facilitate accessible and imaginative storytelling in the realm of video games.
publishDate 2023
dc.date.none.fl_str_mv 2023-12
2023-12-01T00:00:00Z
2024-02-28T15:31:57Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10362/164247
url http://hdl.handle.net/10362/164247
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron:RCAAP
instname_str Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron_str RCAAP
institution RCAAP
reponame_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
collection Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository.name.fl_str_mv Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
repository.mail.fl_str_mv
_version_ 1799138177155334144