关键词: 碰撞检测, 混合包围盒, 任务结构, 三角形相交, 坐标系变换, 降维

Abstract: In the problem of collision detection, the speed and accuracy is one of the key challenges for many computer application programs. A new detection algorithm is proposed that can improve the detection speed and the accuracy, it is based on two phases：Firstly, in the preprocessing detection stage, uniformly dissect the space which is measured to determine the adjacent objects and constructs the AABB-OBB mixed hybrid bounding box. The structure of the bounding box and the task structure are optimized which accelerate the ergodic process; then, in detailed test phase, a new calculation of coordinate system based on M?ller algorithm is improved. The space geometric triangle is projected and reduced aiming at the space problems in two-dimensional plane, so that the total calculation of the algorithm is reduced. Compared with the traditional algorithm, the experimental results show that the detection speed of the new algorithm is greatly improved under the premise of ensuring the accuracy of collision detection.

Key words: collision detection, hybrid bounding box, task structure, triangle-triangle intersection, coordinate system transformation, dimension reduction