|
This article will assume a basic understanding of the geometry and math involved in collision detection. At the end of the article, I’ll provide some references in case you feel a bit rusty in this area. I’ll also assume that you’ve read Jeff Lander’s Graphic Content columns on collision detection (“Crashing into the New Year,” ; “When Two Hearts Collide,”; and “Collision Response: Bouncy, Trouncy, Fun,” ). I’ll take a top-down approach to collision detection by first looking at the whole picture and then quickly inspecting the core routines. I’ll discuss collision detection for two types of graphics engines: portal-based and BSP-based engines. Because the geometry in each engine is organized very differently from the other, the techniques for world-object collision detection are very different. The object-object collision detection, for the most part, will be the same for both types of engines, depending upon your current implementation. After we cover polygonal collision detection, we’ll examine how to extend what we’ve learned to curved objects. |
|