الفهرس 
1.1 Game DevelopmentOnly 14 pages are availabe for public view
 |  | from | 32 |  |  |
| | | from | 32 | | |
Abstract
Real-time Strategy game (RTS) is a war simulator where several opposing factions battle in a virtual world. There are a lot of the challenges concerning this type of games. The most important challenge facing developers is how Game Artificial Intelligence is handled in this genre. There are great efforts done to build a reliable Game AI for most RTS Games. There are attempts done to improve the game performance by investigating intelligent movement of a group of units. This thesis includes an introduction of what group AI and investigates which methods this concept covers. This thesis involves two essential parts, the first part is flocking which dictates how each entity moves in reaction to the others. Flocking behavior is a natural phenomenon which is a convenient method to simulate natural and intelligent group movement in computer games. The ability to maintain flocking (or group) motion for multiple moving objects is an important subject in computer science. In this direction, the algorithms typically do not have to deal with complex environment. Applications of flocking motion include simulating groups of artificial life, or animating troop movements in real-time strategy games. The second part is enhancing the flocking using genetic algorithm (GA). The moving vector, resulting from the flocking behavior, is a linear combination of every simple behavior rule vector. The moving vector weights are optimized to have a realistic flocking moving behavior by using genetic algorithm. In our thesis, there is an investigation and analysis of the path-finding using flocking behavior of groups in real-time strategy games manifested by an open source RTS engine and game. Furthermore, there is utilization of the information sharing between entities in a group, in order to reduce the time for path finding and improve flocking behavior. The proposed technique aims to improve unit movement and decrease unit lose in real-time strategy games. This technique has been applied in some situations that its measurements indicate performance improvement concerning unit damage. For evaluation of this method, a complete, mature and commercially comparable open source RTS game called Wargus is used. This is done by comparing the performance of the flocking behavior with the performance of normal behavior in Wargus. The experiments performed in this thesis show that there is room for improvement in the way unit movement are handled. This technique has been applied in some situations that its measurements indicate improvement of the game performance concerning unit damage.