الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 93 |  |  |
| | | from | 93 | | |
Abstract
The research aims to create better high-level winning of real-time strategy games so; we have presented an experience exchanging model to improve the performance of RTS game engines through exchanging experiences of facing new un-programmed opponent scenarios with each other. Our model is based on the game case-based reasoning system specially on adding new cases to it. New cases are sent by other engines that faced new opponent scenarios and beat them to help the engine dealing with these scenarios if it faces them in the future. The Real-See model is not concentrating on the case description only but on the case goal and performance too. We presented two versions of the system, Real-See [RS] and Real-See Situation Assessment [RSSA]. Both versions were tested in two RTS games Glest and Boswars. Many players preferred the second version [RSSA] as it removes stored cases redundancy, game boredom and provide new challenges to gamers, while others preferred RS to keep scenarios they are familiar with. Simulation results show that using the proposed model as an experience exchanger system has a great impact on improving RTS games abilities of wining and facing new opponent scenarios.