Dynamic difficulty adjustment based on an improved algorithm of UCT for the Pac-Man Game

作者：Bin Wu, DingDing Chen, Suoju He, Qijin Sun, Zhengjun Li, Minxi Zhao · 年份：2011 · DOI：10.1109/icecc.2011.6066649 · 被引用次数：3 · 研究领域：Artificial Intelligence in Games、Reinforcement Learning in Robotics、Digital Games and Media

One major problem of Pac-Man Game is how to generate intelligence of opponents at appropriate challenge level This paper presents an approach for creating game opponents by DDA (Dynamic Difficulty Adjustment), which is used to dynamically adjust challenge level of opponents according to the player's skill level The approach of DDA is based on "time constrained-CI". Furthermore, in order to improve the performance of Pac-Man Game AI, this paper also discusses an alternative approach, "knowledge-based-CI-controlled-NPC", which creates DDA based on knowledge from DTS.