融合GASA的Snake模型碰撞检测算法  

Collision detection algorithm fusing GASA Snake model

作　　者：李照[1] 靳雁霞[1] 秦志鹏[1] LI Zhao JIN Yan-xia QIN Zhi-peng(School of Computer Science and Control Engineering, North University of China, Taiyuan 030051, China)

机构地区：[1]中北大学计算机与控制工程学院,山西太原030051

出　　处：《计算机工程与设计》2017年第5期1223-1228,共6页Computer Engineering and Design

基　　金：山西省自然科学基金项目(2013011017-7);国家自然科学基金项目(61379080)

摘　　要：针对现有碰撞检测算法难以解决变形体碰撞的实时性和逼真性问题,提出一种基于可变形物体的碰撞检测算法。利用Snake模型对FDH包围盒进行更新,由于传统Snake模型存在弱边缘易溢出的问题,结合模拟退火算法(GASA)的遗传思想对其进行优化。实验结果表明,优化后的碰撞检测算法可快速准确得到碰撞物体的最佳轮廓曲线,弥补了遗传算法和模拟退火算存在的过早收敛和低收敛速度的问题,加快了包围盒的更新速度,提高了碰撞检测效率和逼真程度。In view that the existing collision detection algorithm can not solve the real-time and realistic problem of deformation object collision, a collision detection algorithm based on deformable objects was proposed. Snake model was used to update the FDH bounding box. Because the traditional Snake model has the problem of weak edge spillover, simulated annealing （GASA） genetic ideas were combined to optimize the Snake model. Experimental results show that the optimized collision detection algo-rithm can quickly and accurately get the best contour of the object, which makes up for the defect of genetic algorithm and simu-lated annealing algorithm including premature convergence and low convergence speed, accelerating the speed of updating the bounding box, while improving the efficiency of collision detection rate and the degree of realism.

关 键 词：碰撞检测 固定方向凸包(fixed direction hull FDH) 遗传算法 模拟退火 SNAKE模型 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]