Artificial Intelligence Depot
Visiting guest. Why not sign in?
News, knowledge and discussion for the AI enthusiast.
FEATURES COMMUNITY KNOWLEDGE SEARCH  
Understanding Artificial Intelligence
Comprised of critically acclaimed essays by the world's leading experts on each topic in the series, these collections will become definitive texts on crucial issues of our technological times.
More information at Amazon US UK

Reply to Message

Not registered yet?

The AI Depot has a focused community of friendly users. Rather than let anyone abuse the site at the brink of promiscuity, we prefer to let only those with an active interest participate... this simply requires registering.

Why not sign up!

Joining the site's community is completely free. You can then post messages freely, and customise your personal profile at will. Specific privileges will also be granted to you, like being able to access printer-friendly articles without restrictions. So, why not register?

Username:
Password:
Subject:
Email me when someone replies.
Body:

Parent Message

Navigation Research

James,

Welcome to the message boards.

YES! It would make a valid research project. In fact, my 4th year masters research is partly based on what I discuss in my column ;) It's very interesting stuff, and you can get a potential supervisor interested by mentioning the links with robotics -- keeping the game AI thing in the back of your mind! That said, some universities are more tolerant than others, and a game engine is a great platform.

However, there are some things you should know. It's a fair bit of work! I've been working on my navigation for nearly 11 months now, including about 5 quite seriously (almost full-time). Admittedly, most of that was about gaining experience for me, and looking back, I'm a bit disappointed I've taken so long... but my project considers a full 'automated navigation system', so you can break that down until you feel satisfied with the amount of work in front of you (and you can always tackle more if you complete chunks successfully ;)

The other thing you need to consider is that there's a lot of reading involved. The development is only a small part of the thesis, and you can expect at least 60%/70% of the work to be about acquiring knowledge and reading about existing work. Some of that will help your implementations a fair bit though!

There's more and more information out there on virtual navigation (i.e. non-robotic), since game AI developers are becoming more and more open. But it'd be good to find a good compromise between the two fields.

Sure there's a fair number of other people interested in this, and you can count on me for bouncing ideas off! You know where to find me ;)

When's the deadline?

Alex

935 posts.
Thursday 14 February, 13:12
Reply
AI Research

Hi Alex

Thanks for your interest :) I am glad that I have found someone who may be able to somewhat guide me a bit, since I have a lot that I need to get through it seems :)

I dont think that my university will have much problem with the research being focused on games. My project leader is passionate about AI, so I think he will enjoy something interesting :)

I am well aware of the fact that I am going to be reading most of this year ;) The project may end up being a simple literature survey, since I don't know how much time I will have available to myself to actually do a lot of in depth testing. I would however, very much like to have a go at something :)

The next deadline for me to think about is the project proposal. That has to be in on April 11, 2002. The whole thing must be pretty much finished by Aug, Sep, since we have to work on presenting the project, cos we have to present the project at the end to a bunch of people :p

One of my concerns for the project is finding reliable sources. Since the project is an academic paper, the references need to be reliable. I am a bit worried that I will not find that much 'published' content on game AI etc. I am busy going through all of your links and stuff, they seem pretty good :)

One last question. I am studying full time at the moment, so time is going to be a factor in this project. Do you think that I should just do an investigation into the different approaches to AI in games, or should I go more indepth and focus on one problem, the way that u have?

I spose that will depend on how quickly I pick this stuff up, and if I have time ... anyway let me know what u think

Thanks a lot for your time! :)

James

3 posts.
Monday 18 February, 14:43
Reply
Game AI and Research

I can be difficult combining the game AI and research, but once you know how, finding the right balance is easy and fun.

Most game AI coders are stuck in their own little world, and stick to the same algorithms. If it wasn't for game AI academics, in 10 years we'd still have agents doing just A* and FSM. It's your job to pick the cool algorithms from academia, and apply them efficiently to a game environment.

For navigation, the field of reference is robotics. Personally, I've already started my background review, and most of that is pure robotics. Of course, I mention some commercial game AI solutions, but simply to justify the need for my (hopefully better ;) approach ;)

Focusing on lots of small problems can be fun, but not extremely rewarding. Tackling one thing really well will get you a better project, and most likely a good job later ;)

Let me know if there's anything else!

935 posts.
Friday 22 February, 04:16
Reply
Artificial Life ...

Hi Alex!

I have just finished reading the book "Creation: Life and how to make it" by Steven Grand. I am not sure if you are familiar with his work, but I am sure that you are :) I find his point of view and approach to artificial life and intelligence very insightful, and it addresses many of the problems that I have found with AI so far. I find it interesting though, that I can not find much material on the net relating to his approach of creating artificial life. What are your thoughts on his approach to doing things?

Thanks for your time :)

James

3 posts.
Monday 25 February, 15:36
Reply
Creatures

I haven't read the book, so my knowledge of his work is limited to the few articles I've read on the web.

What Steve has done deserves a lot of respect. He found a niche in the market (consciously or not) for artificial life simulations. His agents, however, are nothing amazing. He's built a well-designed creature that does what it does really well. It can learn some things, indeed, but that does not make it revolutionary in any way (let alone "intelligent"). Some interesting behaviours arise from this simplicity, but then even the simplest agents can have complex emergent behaviours.

I'll give him credit for making such a polished product out of it.

His book is on my list. Though Lord of the Rings is taking me longer than I expected... hehe

935 posts.
Monday 25 February, 20:35
Reply

Back to the Artificial Intelligence Depot.