Another Source

22/04/10

Collision Detection Research Group

Introduction


Virtual environment technologies made it possible to include

users in an interactive, 3D world that actually responds to user interaction in real-time. One of the earliest applications of virtual environment is in the area of training and simulation, significantly in flight simulators. Applications of virtual environment broadened into other fields, such as medical surgical simulator, scientific visualization, virtual prototyping and manufacturing, education, game and robotics to name a few. Each case requires users to interact with a virtual model of some sort, which in turn, requires at least collision detection to make the model reacts correctly. In short, objects’ movement are restricted and effected by collision as well as other dynamics constraints. Virtual models that incorporate physics-based rules may need to be able to handle contact determination and collision response, after collision is detected. Therefore, it is apparent that collision detection is a tangible problem in the fields mentioned above.




The term collision detection refers to a task of determining

whether two or more objects intersect. From the most recent surveys on this field, all collision detection techniques can be grouped or classified into two broad categories, namely: bounding volumes hierarchy which essentially is a successive approximation the object's model and spatial decomposition of the model's space. The main objective of collision detection is to generate or produce a report on any intersection that exists between two or more objects. The actual problem and the degree of accuracy of collision detection algorithm depend on the type of application that incorporates it. A surgical simulator may need a very precise collision detection algorithm, compared to flight or driving simulators where precision is not the utmost essential. Game application, on the other hand, need a very fast collision detection and a lesser degree of accuracy. Therefore, most of the proposed algorithms and techniques cannot be considered as universal or optimal solution to the problem at hand. Different algorithms and techniques can best tackle different issues in different application. As a matter of fact, precision and speed are trade-offs in collision detection.

The CDRG research group can be reached at:

Department of Computer Graphic & Multimedia
Faculty of Computer Science and Information Systems,
Universiti Teknologi Malaysia,
81310 Skudai,
Johor Bahru,Malaysia


E-mail : cdrg_gmm@yahoo.com





Website : http://gmm.fsksm.utm.my/~cdrg/home.html#