| Abstract [eng] |
In computer science, including physically based simulations, road planning and in representing the issues in the collision detection is a fundamental problem. Over the past decades, many algorithms have been proposed to speed up collision detection queries. However, there are still a number of open challenges. In order to ensure a fair, different collision algorithms comparison I recommend several methods. This includes theoretical hierarchical collision detection algorithms time operating model and open source compareing which the performance of the collision detection. In this work has studied the performance of algorithms that operate using the CPU (Central processing unit) and GPU (Graphics processing unit), analyzing spatial objects conflict detection methods. The work objectives achievement is realized in the base version of algorithms have been integrated into a single package in order to specifically allow for comparison all algorithms time. This paper studies have shown that there is no single tool suitable for all scenarios. However, the generic would be readily offer Dop-Tree, as in his case combines close BV and quick BV tests. Another study showed that while increasing the complexity of the subject, the collision detection time is improving, but it's just outstanding options. |
