Genres are used to categorize video games according to the way players interact with them as well as their rules.
Although each video game is unique, they can share a number of concepts depending on their genre. This makes it difficult to create thoroughly robust AI because its development is constrained to the scope of an individual game project. On the other hand, AI is usually independently designed for each game. Thus, creating a truly smart and fully autonomous agent for a complex video game can be as challenging as replicating a large part of the complete human intelligence. The richer and more complex a game is, the more skills and abilities it requires. Since video games are designed for human beings, it is only natural that they focus on their cognitive skills and physical abilities. Thus, the scope of discussion is limited to the game aspect in this work.
Conversely, context AI would deal with context-specific tasks such as making a character perform a series of actions to advance the plot or reacting to player choices. This work focuses on game AI, that is, AI which is concerned with solving the problems in the game such as defeating an opponent in combat or navigating in a maze. On the other hand, the context encompasses all the elements that make up the setting in which these problems appear, such as characters and plot. The game includes the elements that define the actual challenges players face and the problems they have to solve, such as rules and objectives. A video game can be considered to have two main aspects, the context and the game. Introductionīecause artificial intelligence (AI) is a broad notion in video games, it is important to start by defining the scope of this work. The approach is illustrated using two video games,
Which relies on conceptual views and actions toĭefine basic yet reasonable and robust behavior. To enable the development of conceptual AI Inspired by the human ability to detect analogiesīetween games and apply similar behavior on aĬonceptual level, this paper suggests an approachīased on the use of a unified conceptual framework Genre, but of different genres too, resulting in aĭifficulty to handle the many aspects of a complexĮnvironment independently for each video game. One issue with this approach is that it does notĮfficiently exploit the numerous similarities thatĮxist between video games not only of the same To quickly and efficiently create specificĪI. Tools currently focus on allowing video game developers Specifically designed for each game, video game AI The various aspects of complex environments With modern video games frequently featuring sophisticatedįor smart and comprehensive agents that understand