Tag Archives: Artificial Intelligence in Games

AI and Games – YouTube Channel

Since my time at The University of Edinburgh, one of my interests has been A.I in games. One great resource I came across that is both informative, and accessible is a YouTube Channel called AI and Games.

The channel is hosted by Tommy Thompson, an Artificial Intelligence researcher, computer science lecturer, and game developer. The channels videos/lectures, and accompanying articles on their site make useful materials for anyone interested in the field of AI and Games.

AI and Games
AI and Games

I’ve personally found these resources to have been most helpful. In particular an informative video that had a look at the A.I. used in the game F.E.A.R helped me write my notes on the A.I. employed by F.E.A.Rs NPCs.

If AI and Games is also one of your areas of interest, be sure to give this a channel a look! To kick things off, here’s a video by Tommy Thompson which serves as a good place to start.

https://www.youtube.com/watch?v=MTKJaN9jwbg

FEAR’s NPC A.I.

FEAR has been cited as one of the most influential A.I. games. FEAR’s claim to fame was its implementation of NPC A.I. using a system called Goal Oriented Action Planning.

Since I have some experience with planning in games, I wanted to have a look at the techniques used. After some digging I thought I’d put together a brief summary of what I learnt.

fear_logo

Continue reading FEAR’s NPC A.I.

Planning in Games

At University my final year project involved building a plan recognition agent for the board game R.I.S.K. Since then, reading about plans in games is a past time of mine, and AiGameDev is a great place to do that.

planning-in-games-aigamedev-logo

For those of you with an interest in A.I. in games, AiGameDev is one of the best resources I’ve come across. Whilst browsing through their impressive collection of articles I came across a gem on planning in games written by Alex J. Champandard.

Check out the full article here!

Reading this article and watching this video was a joy. It reminded me of nights spent learning about STRIPS and HTN’s from Game Developer magazine. So if your interested in planning in games, this is definitely a great place to start.

The Creature A.I. of Black and White

Black and White is where my interest in A.I. in games started. (A little context) Black and White is a god game designed and created by Lionhead Studios under the directive of Peter Molyneaux. In Black and White you are a god born of the prayers of people, free to do whatever you please through the islands of the game.

In your journeys you are eventually given a creature. A tiny little thing, it behaves just like a child – curious yet inexperienced and frightened of the big bad world. Similar to a child you can teach it, it will then think and act based on what  was taught by you. Invest enough time and your creature can become your ultimate agent. It was a fascinating example of A.I. in games.

black-and-white-creature
The Creature Cave

Black and White’s Artificial Intelligence was designed and implement by Richard Evans, and in my opinion it was the creature feature in particular that elevated the experience of certain aspects of the game. Employing sophisticated A.I techniques, it gave deeper meaning to your relationship with the creature, you felt responsible for it (well at least I did) having to train and look after it while it grew up.

Notable features related to the creature included:

  • Leashes – Which gave the player some behavioral control of the creature.
  • Creature Combat – Where creatures would fight autonomously when not commanded by the player.

A battle from Black and White: Creatures Isle. The Crocodile is controlled entirely by the game.

AiGameDev, who listed it as the most influential AI game, summed up the technical innovations as:

  • The gameplay is focused on the interaction with a large AI creature which can learn from examples, and takes reward and punishment.
  • The design integrates artificial life within the context of a strategy game.
  • The engine uses a solid AI architecture, rooted in cognitive science, known as belief-desire-intention (BDI).
  • Machine learning techniques such as decision trees and neural networks are used with great success.

If you are interested in learning more about the innards of Black and Whites creature A.I. then check out this great paper by James Wexler.