oh..this one is a great help..thanks sir. =)
________________________________ From: Thomas Hruska <[email protected]> To: [email protected] Sent: Saturday, July 11, 2009 6:28:27 AM Subject: Re: [c-prog] AI gaming SVMAYOL wrote: > thanks sir,..i understand that you are suggesting a devide and conquer > technique..but im a bit confused on how to converge those which you suggest > (1,2,3,4) and the P.E.A.S...can you show me an example so i might as well > compare.. > > regards, > sv I don't have any examples. A.I. is a bogus term which basically means "wing it until you have something that passes for reality". People typically think "neural networks" when it comes to A.I. but neural networks are simply statistical engines that have been trained on a limited set of data and are only semi-useful - the more inputs into a neural net, the less useful it becomes. http://en.wikipedia.org/wiki/PEAS Performance: One player has the ball. How many times were the other players open to receive the ball? Did all the players stay in bounds? Did the ball stay in bounds? Are your "agents" chasing the ball on its way out of bounds? Are the agents running into each other? (If so, do they bounce off or pass through like ghosts?) When you get to #3 and #4, here are more questions for performance analysis: How many times did they receive the ball successfully (success/failure rate)? How many times did they pass successfully? How many times did they make it into the basket when shooting? How many rebounds did they get when missing the basket? Environment: Players on the basketball court. Partially observable (line of sight passing, shooting, etc.), episodic, dynamic, continuous, multi-agent. Actuators: Uh...the stuff you are displaying on the screen. Could be 2D. Could be 3D. Doesn't really matter if this is a school project. Sensors: Basically a part of Performance. You aren't building a robot and all you probably need are some global variables to do the tracking ("sensing"). My point is, basketball is a large problem space. If you picked this project, then have fun with it and get something basic done right away such as the "Actuators" (a silly name) - display stuff on the screen that shows you have the basic data structures you will need set up right - no movement or anything necessary ('Basketball' , 'Player', 'Court', 'Basket'). Then work on movement (can be completely random at first). Small steps. I'm not here to do your work for you. In terms of PEAS, only Performance really matters. The EAS part is merely an academic exercise that is there just to make sure you know how to logically deconstruct a problem into its component pieces and then write a paper on it. A waste of time if you already know how to do such things. If you are truly stuck, then come up with a different project and ask to do that instead. (Obviously get approval to change projects). -- Thomas Hruska CubicleSoft President Ph: 517-803-4197 *NEW* MyTaskFocus 1.1 Get on task. Stay on task. http://www.CubicleSoft.com/MyTaskFocus/ [Non-text portions of this message have been removed]
