AI Game Engine Programming

by Brian Schwab, Charles River Media (Manufactured by)

AI Game Engine Programming provides game developers with the tools and wisdom necessary to create modern game AI engines. It takes programmers from theory to actual game development, with usable code frameworks designed to go beyond merely detailing how a technique might be used. In addition, it surveys the capabilities of the different techniques used. In addition, it surveys the capabilities of the different techniques used in some current AI engines, and covers common pitfalls, design considerations, and optimizations. If you're having difficulty determining which techniques to use, or looking for working code best suited to a particular game, you'll find the answers here. You'll also find a clean, usable interface for a variety of game AI techniques with an emphasis on primary decision-making paradigms. The book provides insightful coverage of a variety of subjects important to AI engine development, and ties them together masterfully to form an indispensable reference. Part I provides an overall look at game AI, covers the basic terminology used in the book, reviews underlying concepts of game AI, and dissects the parts of a game AI engine. Part II covers specific game genres (RPG, TRS, FTPS, Shooter, Sports, Racing, Strategy, Adventure, Fighting, Platform, and miscellaneous) and explains how they use different AU paradigms. It also covers the move common solutions to the problems posed by each genre. Some of the problems include, dealing with direct AI and human interactions, using scripting languages appropriately, and general intelligence/entertainment balancing. Part III provides the actual code implementations for the basic AI techniques such as finite state machines, fuzzy state machines, message board systems, scripted systems, and location-based information systems. And, Part IV covers the move advanced techniques, including genetic algorithms, neural networks, artificial life, planning algorithms, and decision trees. The book concludes wit

Genres: Programming, Artificial Intelligence, Reference, Nonfiction

594 pages, Paperback

First published September 1, 2004

Reviews

[Chris Rittelmeyer]

A fully revised update to the first edition, "AI Game Engine Programming, Second Edition" provides game developers with the tools and information they need to create modern game AI engines. Covering the four principle elements of game artificial intelligence, the book takes you from theory to actual game development, going beyond merely discussing how a technique might be used. Beginning with a clear definition of game AI, you'll learn common terminology, the underlying concepts of AI, and you'll explore the different parts of the game AI engine. You'll then take a look at AI design considerations, solutions, and even common pitfalls genre-by-genre, covering the majority of modern game genres and examining concrete examples of AI used in actual commercial games. Finally, you'll study actual code implementations for each AI technique presented, both in skeletal form and as part of a real-world example, to learn how it works in an actual game engine and how it can be optimized in the future. Written for experienced game developers with a working knowledge of C++, data structures, and object oriented programming, "AI Game Engine Programming, Second Edition" will expand your AI knowledge and skills from start to finish.

[Anthony L · Rating 1 out of 5]

I found this book at a book exchange corner. I happen to have a postgraduate research degree involving neural networks (NN) . Naturally, I am curious what this book has to say in the chapter on NN, even though it was published so many years ago. I am shocked that the chapter is completely WRONG in the way NN are deployed. It talked about NN being used for regression, looking like an awkward academic-like discussion rolled out by someone obviously with no real background in NN deployment . NN are used for pattern recognition which it mentioned in passing as being used by the postal service for recognising handwriting.

The critical transformation of input into the form of binary data needed for a NN is missing. Instead, for the codes, it simply said there are four inputs, none of which are the binary data for NN inputs! A neuron, whether input or otherwise either fires or does not fire. No way an input neuron can take speed or whatever directly as input! The entire basis is wrong even though the cursory explanation of what are NN is there, which is nothing much in this day of googling.

[Marciano Viereck · Rating 4 out of 5]

Pretty good book & useful reference