Collision Detection Strategy for Interactive Virtual Environment

N.M. Suaib, A. Bade, D. Daman, and M.S. Sunar (Malaysia)


Collision detection, interactive virtual environment,computer graphics.


This paper will introduce a review of some of the previous work in collision detection in Section 2, focusing on bounding volumes hierarchies. In Section 3, we will describe our unified bounding volume strategy, construction of the AABB and k-Dops hierarchy, along with the traversal algorithms. We conclude this paper with our current research directions in order to perform real-time and faster collision detection scheme. This paper presents a collision detection strategy for interactive computer simulated environment. The environment consists of large and complex dynamic objects undergoing rigid-body motion. The strategy relies on two distinct portions. In order to achieve fast overlap checking, we perform Hierarchical Axis-Aligned Bounding Box (AABB) in the first phase. Once AABB testing is positive (indicating detected collision), Discrete Orientation Polytope (k-Dops) approach will then be conducted as a sureness indicator for collision detection checking. We believe that our potential strategy is appropriate for large and complex dynamic environment. 2. Previous Work Many computer graphics algorithms use Bounding Volume (BV) approach in order to improve their performance. The BV of a geometric object is a simple volume enclosing the object. BVs are used to speed up computation and rendering process, but they are not directly involved in the rendering process [16]. BV has long been used in other fields. The primary exploration was probably used in raytracing algorithm since 1986[15]. Some literatures claimed that earlier effort using BV for collision detection schemes was set up by Moore and Wilhems in 1988 [9].

Important Links:

Go Back