Hello
I'm working on an OPENSOURCE Soccer Game in C++
and I'm interested in particular on the AI.
I'm intended to create a Google Group just about this topic...
In this way we can share our experiences and suggestions..
You are wellcome.
soccer game movement help
November 23, 2006 09:12 AM
You all are welcome
This is the name: soccer.ai.programming
http://groups-beta.google.com/group/socceraiprogramming
The group is brand new now...
come in and post there
This is the name: soccer.ai.programming
http://groups-beta.google.com/group/socceraiprogramming
The group is brand new now...
come in and post there
Quote: Original post by Anonymous Poster
Hello
I'm working on an OPENSOURCE Soccer Game in C++
and I'm interested in particular on the AI.
I'm intended to create a Google Group just about this topic...
In this way we can share our experiences and suggestions..
You are wellcome.
What's wrong with discussing this topic within this forum? The benefit you'll have over a topic-specific group list is that there are often a lot of lurkers in this forums with very detailed knowledge of certain techniques useful in AI. If you take your discussion away from this thread, you will generally lose the opportunity for these people to chime in with important information when it could do some good.
Cheers,
Timkin
Quote: Original post by TimkinQuote: Original post by Anonymous Poster
Hello
I'm working on an OPENSOURCE Soccer Game in C++
and I'm interested in particular on the AI.
I'm intended to create a Google Group just about this topic...
In this way we can share our experiences and suggestions..
You are wellcome.
What's wrong with discussing this topic within this forum? The benefit you'll have over a topic-specific group list is that there are often a lot of lurkers in this forums with very detailed knowledge of certain techniques useful in AI. If you take your discussion away from this thread, you will generally lose the opportunity for these people to chime in with important information when it could do some good.
Cheers,
Timkin
this is true, I've enjoyed my short time on this forum and have found everyone extremely helpful and full of good ideas! I'm hoping interest in this thread will continue and people enjoy posting their ideas on the subject.
today I'm contemplating how to implement goalkeeper AI - with attributes such as Handling, Reflexes, Shot Stopping, etc.
I'm attempting to get more realistic movement for the players - I have just been reading articles on Newtonian physics and linear algebra but none seem to be helpful to me - anyone know of any resources online that could be applied to what I'm trying to achieve?
November 28, 2006 01:26 PM
Quote: Original post by Timkin
check out Penny Sweetser's homepage at UoQ.
Wow! Smart and hot... I need to meet this girl.
could someone help me out with a problem I'm having?
I want to get my players to make curve/swerving passes rather than straight line passing all the time - how do I go about doing that?
I want to get my players to make curve/swerving passes rather than straight line passing all the time - how do I go about doing that?
You could use a bezier or hermite or any type of quadratic curve. I personally like this one because it's easy to use:
X = (2*c-4*b+2*a)*t*t+(4*b-c-3*a)*t+a ;
It goes through a at t = 0, b at t = 0.5, and c at t=1 . If you can define 3 control points and then scale your actual time to be in the 0 to 1 range it is quite easy to use. If you know when you want your pass to end(the total time your pass will take) then you can define your final point as reciever position+ reciever velocity*time in the same way as a linear pass. Of course there is more than one way you could do this depending on what kind of result you're trying to get.
edit: hermite curves are cubic, and bezier curves can also be cubic or higher...just clearing that up because the way I said it kind of makes them sound like they are just quadratic
[Edited by - Alrecenk on December 1, 2006 1:20:10 PM]
X = (2*c-4*b+2*a)*t*t+(4*b-c-3*a)*t+a ;
It goes through a at t = 0, b at t = 0.5, and c at t=1 . If you can define 3 control points and then scale your actual time to be in the 0 to 1 range it is quite easy to use. If you know when you want your pass to end(the total time your pass will take) then you can define your final point as reciever position+ reciever velocity*time in the same way as a linear pass. Of course there is more than one way you could do this depending on what kind of result you're trying to get.
edit: hermite curves are cubic, and bezier curves can also be cubic or higher...just clearing that up because the way I said it kind of makes them sound like they are just quadratic
[Edited by - Alrecenk on December 1, 2006 1:20:10 PM]
Alrecenk, thanks for that! that works for the Y-axis but doesn't seem to work for the X axis, I've been trying to figure out what I need to do to get the X-axis to work too unsuccessfully so far. how do I do it?
edit: I've figured out the problem - it's when point C is less than point B that there's a problem.
I tried a curve with the x positions of 25, 50, 40 and it's not working - I'm obviously doing something I shouldn't be. is it possible to make a curve with co-ordinates like that?
[Edited by - anti_hero on December 4, 2006 12:19:28 PM]
edit: I've figured out the problem - it's when point C is less than point B that there's a problem.
I tried a curve with the x positions of 25, 50, 40 and it's not working - I'm obviously doing something I shouldn't be. is it possible to make a curve with co-ordinates like that?
[Edited by - anti_hero on December 4, 2006 12:19:28 PM]
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement