Object-oriented parallel collision detection method, device and system

Abstract

The invention discloses an object-oriented parallel collision detection method, device and system. A list of all object pairs requiring collision detection is generated by a master control node according to a mobile list; an idle node sends out a request to the master control node to make information of collision detection, the master control node distributes object pairs in the list of all object pairs requiring collision detection to the idle node to carry out parallel collision detection after receiving the request to make the information of collision detection. The method, device and system can also lower the arithmetic complexity and further increase the collision detection speed while ensuring the uniqueness and the completeness of collision response.

Description

technical field [0001] The invention relates to the field of collision detection, in particular to an object-oriented parallel collision detection method, device and system. Background technique [0002] The problem of collision detection has been concerned for many years. The collision detection method has many applications in the fields of fabric simulation, computational geometry, computer animation, CAD / CAM, simulated robot and virtual reality. In the past two decades, researchers have done a lot of work in the field of collision detection and proposed some relatively mature algorithms. [0003] In general, these algorithms can be divided into two categories as a whole: geometric decomposition methods and hierarchical bounding box methods. [0004] The core idea of ​​the geometric decomposition method is to divide the entire virtual space into small cells of equal volume, and only calculate the intersection of geometric objects occupying the same cell or adjacent cells....

AI Technical Summary

Problems solved by technology

Hierarchical bounding box technology is used to speed up the collision detection of polyhedral scenes, but the bounding box method as a whole method has low detection accuracy and the efficiency improvement is not obvious

[0008] (2) Repeatability of detection: in the prior art, when performing collision detection on moving objects, each moving object will perform a collision detection with all objects in the environment. In this way, for a pair of moving objects, both parties Each collision detection is done once, resulting in the repetition of a collision detection, increasing the number of collision detection

In order to solve this problem, some other measures have to be taken to deal with the uniqueness of the collision response: before each collision response, it is necessary to determine whether the collision has been processed, which undoubtedly increases the complexity of the algorithm

[0010] (4) In the prior art, the update of the object state is regarded as a member function of the object (that is, the object). There are many kinds of motion equations. If the motion update is completed by the member function of the object, the program lacks a general interface. Every time the motion equation is modified, the member function must be modified. Various forms of motion equations will update the state of the object. kind of hard

Images