Collision Detection Algorithm For Virtual Surgery

Collision detection is one of the key techniques in virtual environments, and its speed and accuracy greatly influences the reality and sense of immersion. With the background of virtual surgery simulations, the collision detection and self-collision detection problems are studied comprehensively in this thesis.Firstly, a speedy hybrid hierarchical bounding volume algorithm (SHBVs) based on AABB and OBB is proposed, which may combine the quality of all single hierarchical bounding boxes, and overcome the shortages in real-timing and accuracy. By analyzing the characteristics of various hierarchical bounding volumes and the objects in virtual environments, the bounding volumes tree is divided into two levels:on the top level,AABB-AABB is used as a tree node to eliminate the possibility of collision between objects far apart, and on the lower level, OBB-AABB is used to bound the moving object tightly by OBB and bound the environment object for updating the bounding volume quickly after distortion by AABB.Experimental results demonstrate that the proposed algorithm works well.The spatial subdivision algorithm is used in virtual surgery when the soft objects collide themselves.We properly extend the space area when dividing the units, through enlargering space area to reduce time and increase the efficiency.Hash list is used to store the results when mapping the objects to units.The algorithm can meet the real-time quality in virtual surgery.Finally, this paper implements the collision detection and self-collision detection for virtual surgery.The experiment results shows that it can solve the problems and fulfil the requirement of accurate and real-time collision detection in virtual surgery.