Publication: Un modelo computacional para representar videojuegos multijugador de detectives e implementar el comportamiento de participantes automáticos
Loading...
Official URL
Full text at PDC
Publication Date
2016
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citations
Exportar
Abstract
En la actualidad, los videojuegos no llegan a alcanzar el nivel de realismo esperado debido, entre otros factores, al trato que se hace del conocimiento de sus personajes. En muchas ocasiones estos personajes no son capaces de actualizar su conocimiento acerca de lo ocurrido convenientemente, dando lugar a situaciones extrañas. Este trabajo aborda el problema de la gestión de conocimiento, el razonamiento y la comunicación entre personajes de videojuego controlados por computador, algo que la industria del entretenimiento interactivo está muy interesada en conseguir para aumentar la jugabilidad y la credibilidad de sus obras. Para explorar las posibilidades de construir personajes con una inteligencia artificial capaz de extraer conocimiento sobre lo que perciben en su entorno, y propagarlo a otros de manera realista, imperfecta o incluso malintencionada, se propone un escenario ficticio propio de un videojuegos multijugador de detectives, donde además de participantes humanos, vamos a explorar la posibilidad de implementar participantes automáticos. El escenario, de temática fantástica y humor, consiste básicamente en un laberinto de un laboratorio donde un grupo de ratones pueden moverse y realizar acciones por la noche. A la mañana siguiente el científico del laboratorio inspeccionará el estado del laberinto e intentará averiguar ratones han causado desperfectos para castigarlos. En este trabajo explicamos el proceso llevado a cabo para modelar este juego y representar computacionalmente el conocimiento y los pasos de razonamiento que deben dar los participantes para jugar de forma plausible al mismo. Con este modelo sentamos las bases de un armazón que permita explorar diferentes estrategias de resolución de problemas ante toda una familia de escenarios posibles donde hay personajes que intercambian información tratando de maximizar su puntuación en el juego. Finalmente se ofrece una solución al problema con ratones que son conscientes de las acciones que pueden acarrear consecuencias que han realizado y por lo tanto mienten en sus declaraciones para evitar ser castigados. Los ratones inocentes pretenden sacar a la luz la verdad. El científico sigue un método de investigación que le lleva a dar con un ratón sospechoso.
Nowadays, videogames do not achieve the expected level due to the treatment that is done to the knowledge of the characters. Usually, this characters cannot update their knowledge about what happened correctly, provoking strange situations. This work takles problems related to knowledge management, reasoning and characters comunication in videogames, something that this industry is very interested in achieving in order to improve its gameplay and the credibility of its works. In order to explore all the possibilities of creating characters whose artificial intelligence can get the knowledge of its enviroment and propagate it to others in a realistic, imperfect or even malicious way, we propose a ficticious stage of a multiplayer videogame about detectives, where moreover characters led by the player, we are going to explore the possibility of introducing non player characters. The stage, which is about a fantastic and funny topic, is mainly a maze in a laboratory where a group of mice can move a carry out actions during the night. In the morning, the scientist of the laboratory will check the maze conditions and he will try to discover which mice made damage to punish them. In this work we explain the process carried out to shape this game and the process to represent the knowledge and the reasoning steps that the characters must follow in order to play in a right way to the same one. This model sets the foundations of a framework that allow us to explore the different strategies to solve problems about possible stages where the characters exchange information and try to maximize their punctuation in the game. Finally, we propose a solution to the problem that consist in a group of mice conscious of the consequences of their actions so they decide to lie in their declarations to avoid be punished. The innocent mice try to show the truth while the scientist follows an research method to know who is the guilty mouse.
Nowadays, videogames do not achieve the expected level due to the treatment that is done to the knowledge of the characters. Usually, this characters cannot update their knowledge about what happened correctly, provoking strange situations. This work takles problems related to knowledge management, reasoning and characters comunication in videogames, something that this industry is very interested in achieving in order to improve its gameplay and the credibility of its works. In order to explore all the possibilities of creating characters whose artificial intelligence can get the knowledge of its enviroment and propagate it to others in a realistic, imperfect or even malicious way, we propose a ficticious stage of a multiplayer videogame about detectives, where moreover characters led by the player, we are going to explore the possibility of introducing non player characters. The stage, which is about a fantastic and funny topic, is mainly a maze in a laboratory where a group of mice can move a carry out actions during the night. In the morning, the scientist of the laboratory will check the maze conditions and he will try to discover which mice made damage to punish them. In this work we explain the process carried out to shape this game and the process to represent the knowledge and the reasoning steps that the characters must follow in order to play in a right way to the same one. This model sets the foundations of a framework that allow us to explore the different strategies to solve problems about possible stages where the characters exchange information and try to maximize their punctuation in the game. Finally, we propose a solution to the problem that consist in a group of mice conscious of the consequences of their actions so they decide to lie in their declarations to avoid be punished. The innocent mice try to show the truth while the scientist follows an research method to know who is the guilty mouse.
Description
Trabajo de Fin de Grado en Ingeniería Informática y Matemáticas (Universidad Complutense, Facultad de Informática, curso 2015/2016)