This article written by Andrew Giese
To better understand Artificial Intelligence, we spoke with Michel Kripalani, Director of Business Development, Games Technology Group, who has been involved in projects developing tools like 3ds Max, Maya, HumanIK, and a premier artificial intelligence middleware solution named Kynapse. Kynapse’s touch can be easily recognized in popular games like Crackdown and the upcoming blockbuster Fable 2.
To get a more academic view of how AI functions in today's world, we also got in touch With Dr. Dale Courte, associate Professor of Computer Science at the University of Dayton. Dr. Courte specializes in researching evolutionary computation (technical for “self-perfecting equations”), pattern classification, and machine learning.
Asking the question “What is Artificial Intelligence?” is sort of like asking “What is life?” There are so many different disciplines in the field that it would take textbooks to cover them all. “The phrase 'Artificial Intelligence' is sort of a catch-all for any computing that goes beyond simply automating tasks that used to be done by hand,” says Dr. Courte. “I think of AI as a collection of sub-disciplines, some of which attempt to accomplish, at least to some extent, goals like [data mining, automating complex calculations, and handling machine-work too difficult for humans]."
On the gaming front, Kripalani enlightens us, “Game AI is comprised of three distinct areas; perception, decision and action. Characters need to perceive the world then make a decision and then take action. It is easy to jump straight to the decision making and believe that is what AI is about, but in reality it all starts with perception. Where are the pathways? Where is there a natural cover location? What are the danger areas? In general, we find that many people spend too much time on decision making and not enough on perception. Without the right understanding of your environment you cannot make the right decision.”
Making a program smart enough to understand a problem, like traversing an environment, is a huge obstacle for scientists to overcome.
“It is hard to understand the 3D space…” comments Kripalani. “The whole problem has gotten a lot more complex because worlds are becoming dynamic. The introduction of physics and dynamic worlds is creating environments that can change drastically. Everything can explode, move, disappear, re-disappear, etc. This is where we encounter the largest amount of obstacles in developing Kynapse.
Artificial Intelligence is taking increasingly more important roles in new video games. Instead of the once-simple enemies that patrolled set paths and performed an attack when the player got within an arbitrary distance, gamers of today expect enemies that work together, solve problems, and most importantly, express emotion. A game’s artificial intelligence system is designed such that all these goals are met and enacted realistically in the right context in-game.
To get a better grasp of artificial intelligence in video games, we spoke with the people behind Kynapse, the same A.I. middleware solution used in the upcoming Fable 2.
To get a more academic view of how AI functions in today's world, we also got in touch With Dr. Dale Courte, associate Professor of Computer Science at the University of Dayton. Dr. Courte specializes in researching evolutionary computation (technical for “self-perfecting equations”), pattern classification, and machine learning.
Asking the question “What is Artificial Intelligence?” is sort of like asking “What is life?” There are so many different disciplines in the field that it would take textbooks to cover them all. “The phrase 'Artificial Intelligence' is sort of a catch-all for any computing that goes beyond simply automating tasks that used to be done by hand,” says Dr. Courte. “I think of AI as a collection of sub-disciplines, some of which attempt to accomplish, at least to some extent, goals like [data mining, automating complex calculations, and handling machine-work too difficult for humans]."
On the gaming front, Kripalani enlightens us, “Game AI is comprised of three distinct areas; perception, decision and action. Characters need to perceive the world then make a decision and then take action. It is easy to jump straight to the decision making and believe that is what AI is about, but in reality it all starts with perception. Where are the pathways? Where is there a natural cover location? What are the danger areas? In general, we find that many people spend too much time on decision making and not enough on perception. Without the right understanding of your environment you cannot make the right decision.”
Making a program smart enough to understand a problem, like traversing an environment, is a huge obstacle for scientists to overcome.
“It is hard to understand the 3D space…” comments Kripalani. “The whole problem has gotten a lot more complex because worlds are becoming dynamic. The introduction of physics and dynamic worlds is creating environments that can change drastically. Everything can explode, move, disappear, re-disappear, etc. This is where we encounter the largest amount of obstacles in developing Kynapse.
You must be logged in to post a comment.
 |
|
 Games, News, Reviews, Media and More |
|
 Review: Valkyria Chronicles |
11/17/08 |
 |
|
| Review: Mirror's Edge |
11/14/08 |
|
|
 Review: Mirror's Edge |
11/14/08 |
|
|
| Preview: Killzone 2 |
11/13/08 |
|
|
 Review: Dragon Ball Z: Infini... |
11/12/08 |
|
|
| Review: Legendary |
11/12/08 |
|
|
 Review: Legendary |
11/12/08 |
|
|
  |
|
