Ogre, Physx and OpenAL is used.
Navigation system uses Recast
Steering and decision systems are hand written.
Although the decision system is mature enough(a hierarchical state machine with priorities) the states of the agents are purely written so they do not seem that intelligent.
The agents try to get behind the walls to get cover when they see an enemy. For this i used the sensor grid algorithm which you can read in the ai wisdom 3 book.
With the help of some good looking animations and more detailed agent logic it could be a much more better one.
Download : zzz.rar
Source Code : zzz_source.rar
Short Video : http://www.vimeo.com/12833880
[vimeo]12833880[/vimeo]
