A.I. is ever the hot topic in the industry. Adaptive A.I. is even hotter and seems to be the next big thing and it will be. I think though, personally, Adaptive A.I. will be a progression and it’ll start off small in A.I. and then keep growing and growing. Firstly it’s good that you defined what adaptive A.I. is not: to create such a human like A.I. it’s a lot harder to beat, which is completely not the point as some people expect. So this shows you know what you’re researching at least. The only thing I think I should say is that to keep it pinned down, hard. It’s such a broad topic in the industry right now. Which is good as you’ll be able to find a lot of materials for it. It’s also bad though that you might get lost in it.
Bomberman is quite a good choice to base you’re A.I. code on. A lot of people have played it and know how it plays and a new method of A.I. should be noticeable. Are you going to do testing? Will it involve a spread of people? This will allow you to test your A.I. code in the best possible way and will give you feedback on what people think. Bomberman won’t be easy though as the game progresses the amount of choices available to the player increase tenfold. Yet this won’t be impossible to overcome =using some slecetion techniques.
This, as always, is a good topic to researchc and hope that it will bear fruit for you and I look forward to playing some bomberman.
Monday, 10 November 2008
Worksheet 3
Introduction
The aim of the project is to use genetic algorithms to create A.I. entities capable of altering behaviour in a real time game environment to better compete against players or other A.I. entities, while remaining relatively flexible. Also numerous different variations will be implemented to allow further insight into what works best in a given situation as well as into the issues involved with the methods.
Motivation
Artificial intelligence is becoming more and more important as games attempt to create further realism. By using adaptive A.I. it is possible to have an entity “learn something at runtime from a source of information (such as the person who bought the game) that is unavailable at development time.” (Funge, 2004) Because of this it is possible to create a game entity that can change dependent upon the person playing the game, to better play against the player. It is however important to note that the aim of artificial intelligence is not to create unbeatable opponents but to create “algorithms that make game characters appear human or animal-like.” (Millington, 2006) The motivation behind this project is to explore learning A.I. in an attempt to better learn how best these two elements of A.I. can be brought together to create an A.I. that will learn and react while remaining flexible and not becoming overpowered.
Research Question
How Can Genetic Algorithms best be used to create non-playable characters which are capable of adapting to better compete against different types of player in a real time game environment and what are the issues associated with their use?
Addressing the Question
I will write a multiplayer game, likely to be based on Bomberman, that will contain enough decision making and strategy for the computer controlled players to be able to learn, through playing, which techniques and strategies are the best way to play the game. I will then use several varying techniques to control this A.I. learning and judge them on success as well as flexibility and effect on performance. This will allow me to figure out which methods work best as well as evaluate any issues with the techniques.
Resource Requirements
I only really need a reasonable level PC for the project.
Bibliography/References
John David Funge. (2004) Artificial Intelligence for Computer Games: An Introduction. [Online] Available at: http://books.google.co.uk/books?hl=en&lr=&id=jy_q4URyrowC&oi=fnd&pg=PP13&dq=artificial+intelligence+game&ots=1jzBSq1QLY&sig=8SWQZJMO7W4mS0Io5AkvBwqoy5E#PPP1,M1. [Last Accessed 04 November 2008]
Ian Millington. (2006) Artificial Intelligence for Games. [Online] Available at: http://books.google.co.uk/books?id=zCiv-lMad-AC&dq=artificial+intelligence+game&lr=&source=gbs_summary_s&cad=0. [Last Accessed 04 November 2008]
The aim of the project is to use genetic algorithms to create A.I. entities capable of altering behaviour in a real time game environment to better compete against players or other A.I. entities, while remaining relatively flexible. Also numerous different variations will be implemented to allow further insight into what works best in a given situation as well as into the issues involved with the methods.
Motivation
Artificial intelligence is becoming more and more important as games attempt to create further realism. By using adaptive A.I. it is possible to have an entity “learn something at runtime from a source of information (such as the person who bought the game) that is unavailable at development time.” (Funge, 2004) Because of this it is possible to create a game entity that can change dependent upon the person playing the game, to better play against the player. It is however important to note that the aim of artificial intelligence is not to create unbeatable opponents but to create “algorithms that make game characters appear human or animal-like.” (Millington, 2006) The motivation behind this project is to explore learning A.I. in an attempt to better learn how best these two elements of A.I. can be brought together to create an A.I. that will learn and react while remaining flexible and not becoming overpowered.
Research Question
How Can Genetic Algorithms best be used to create non-playable characters which are capable of adapting to better compete against different types of player in a real time game environment and what are the issues associated with their use?
Addressing the Question
I will write a multiplayer game, likely to be based on Bomberman, that will contain enough decision making and strategy for the computer controlled players to be able to learn, through playing, which techniques and strategies are the best way to play the game. I will then use several varying techniques to control this A.I. learning and judge them on success as well as flexibility and effect on performance. This will allow me to figure out which methods work best as well as evaluate any issues with the techniques.
Resource Requirements
I only really need a reasonable level PC for the project.
Bibliography/References
John David Funge. (2004) Artificial Intelligence for Computer Games: An Introduction. [Online] Available at: http://books.google.co.uk/books?hl=en&lr=&id=jy_q4URyrowC&oi=fnd&pg=PP13&dq=artificial+intelligence+game&ots=1jzBSq1QLY&sig=8SWQZJMO7W4mS0Io5AkvBwqoy5E#PPP1,M1. [Last Accessed 04 November 2008]
Ian Millington. (2006) Artificial Intelligence for Games. [Online] Available at: http://books.google.co.uk/books?id=zCiv-lMad-AC&dq=artificial+intelligence+game&lr=&source=gbs_summary_s&cad=0. [Last Accessed 04 November 2008]
Subscribe to:
Posts (Atom)