The techniques and implementation of fast collision detection for three-dimension pipeline
Zhongliang Fu, Shiwei Shao · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2009
In the design of engineering pipeline, spatial collision detection is always a necessary process. If the artificial method is used, a heavy work load is always needed and omissions occasionally happens; if the method of collision detection algorithm for three dimension object in graphics is used, the computation will increasing sharply with the increase of pipeline number. Because the data size of three dimension pipeline is extremely big, the key to solve collision problem is efficiency. In this paper, a new collision detection method for three dimension pipeline is presented. The new method is composed of two parts: the coarse detection based on Axis-Aligned Bounding Box and precise detection based on geometry model. The method abstracts the pipeline to spatial line-segments with semantic information using geometry method. Firstly, an improved Axis-Aligned Bounding Box algorithm is used to classify the pipelines. Finally, the result of coarse detection is used as initial value, the geometry location relation is used to classify the pipelines, and then the shortest connecting lines between pipelines are calculated. The experiment results show that, this method can detect collision three-dimension pipeline efficiently.