论文部分内容阅读
in traditional computer games,Non Player Characters (NPC) or opponents behavior has been preliminary decided.As a result,players cheat NPC easily by recognizing their fixed behavior and feel boring because of continuous winning.In order to attract players,more intelligent NPC has become the trend of game development.Compared with previous methods,Dynamic Difficulty Adjustment (DDA),a method that adjusts NPCs behaviors according to players behaviors dynamically,provides an available solution to enhance the entertaining of video games.By using this concept,NPC learn to adjust the level of difficulty by " observing" players behavior and tempered difficulty level can make players feel fun during the process of playing.In this paper,we propose to use Artificial Neutral Network (ANN) to implement DDA.By imbedding ANN into test-bed,computer can judge players further behavior and make desired decision after analysis players initial behavior.Different ANN can provide different win rate for different player strategy in order to achieve the dynamic balance we expected and enhance the entertaining of games.