AI Game Development is a bit of a different beast in the AI field, particularly when compared with other game AI references. Although many quite excellent books are on the market, AI Game Development is better than most at providing one core focus: helping the reader learn the techniques to build reactive, learning AIs.
Throughout this book, Alex expands on the concept of what he terms nouvelle game AI, a design concept that combines elements of reaction architectures and learning algorithms. This combination enables developers to develop AIs that are autonomous unto themselves.... They use the same basic algorithms and AI engine functions, but they are independent agents that can act and react to their environment in their own way. These autonomous agents have a natural capability for adapting to (or at least not looking stupid in) new environments, such as the ones found in the ever-unfolding and evolving world of an online role-playing games (RPGs).
These design concepts are sound and represent much of what I've been saying in the industry for many years. The recent popularity of "open-ended" games, such as The Sims, Freelancer, and GTA3: Vice City, show that the game-playing public loves games that "feel" real, which have NPCs and wingmen and passers-by that just "act right."
Meeting those desires are exactly what this book is all about. From the initial chapters on basic navigation behaviors to the more detailed chapters on designing AIs with perception and emotion, AI Game Development provides readers with the tools needed to answer the demands of players over the next few years.