Artificial Intelligence for Computer Games: An IntroductionTaylor & Francis, 29/07/2004 - 146 من الصفحات Learn to make games that are more fun and engaging! Building on fundamental principles of Artificial Intelligence, Funge explains how to create Non-Player Characters (NPCs) with progressively more sophisticated capabilities. Starting with the basic capability of acting in the game world, the book explains how to develop NPCs who can perceive, remember what they perceive, and then continue in the game play to think about the effects of possible actions, and finally learn from their experience. Funge considers the system architecture and explains how to implement potential behaviors (both reactive and deliberate) for intelligent and responsive NPCs allowing for games that are more fun and engaging. Emphasizing enduring design principles, Funge covers the basics of Game AI and provides a clear, easy to read introduction that beginning programmers and game designers will enjoy. |
طبعات أخرى - عرض جميع المقتطفات
عبارات ومصطلحات مألوفة
action choices adversarial search angry animation approximate simulator behavior book's companion calculate Chapter character's controller chase collision computer games const controller's create decision tree defined depth-first search discrete tag game event queue example expert systems Figure first-order logic FSMs function future game actions game AI game designer game developer game world game-state goal graphics grid heuristic high-level implement inline input last tagged learner learning algorithm Listing machine learning memory percept mental state variables move myIndex needs Newtonian physics noisy sensors NPC NPC NPC's controller optimal output parameters particular path planning perception object pick actions pick an action player character possible actions predict predictor percept probability distribution production rules programming random number randomly reactive controller reinforcement learning renderer represent reward search algorithms search tree Section simple stochastic tagged character tagged character's tgCharacter tgReal tion transition model troller unsupervised learning usually values vector velocity waypoint