蒙皮骨骼的碰撞检测技术研究

发布时间：2019-02-09 12:51

【摘要】： 碰撞检测问题在机器人运动规划、计算机图形学等领域中已有很长的研究历史,近年来随着虚拟现实、分布交互仿真等技术的兴起,碰撞检测问题开始成为研究的热点。精确的碰撞检测对提高虚拟环境的真实性、增强虚拟环境的沉浸感有着至关重要的作用,而虚拟环境中自身的复杂性和实时性,又对碰撞检测提出了更高的要求。角色动画是最近十年来计算机科学最活跃的领域之一,已广泛应用于虚拟现实、计算机辅助设计、广告媒体和数字娱乐等诸多领域。随着计算机硬件技术的发展,特别是消费级别的带有硬件加速功能的显卡技术的发展,实时角色动画逐渐获得了越来越广泛的应用。实时角色动画具有巨大的经济前景和理论研究意义。在实时角色动画中,蒙皮骨骼动画是现在使用最为广泛的动画技术。它结合了单一网格动画和关节动画的特点,既克服了单一网格动画需要占用大量内存的缺点,又不会有关节动画时常出现的3D人物身体裂缝问题。本文对蒙皮骨骼动画的碰撞检测技术做出了详细的研究和讨论。在以往的研究中,并没有对蒙皮骨骼动画有专门的碰撞检测研究,而只是把传统的碰撞检测技术运用到蒙皮骨骼动画中,使得针对性不强,不能很好的适用于蒙皮骨骼动画系统。本文在传统碰撞检测技术的基础上,结合蒙皮骨骼自身的特点,改进了碰撞检测算法,以适应蒙皮骨骼动画系统,并通过实验验证了算法的有效性和适用性。在算法的改进上,本文主要做了以下工作: ①利用蒙皮骨骼动画中骨骼结构的特点,选择了OBB作为虚拟角色的包围盒,使得包围盒与骨骼相互联系,成为一一对应的关系。 ②改进了两个子OBB生成父OBB的算法,并在此基础上改进角色包围盒层次树的生成; ③利用蒙皮骨骼动画中Animation结构的特点,对OBB层次树的更新进行了改进,使得OBB的更新能利用虚拟角色的骨骼变换信息,即OBB的变化与骨骼的变换相结合,加快OBB树的更新速度。最后实验证明该算法提高了碰撞检测的效率,能满足虚拟环境中复杂3D人物的碰撞检测需求。
[Abstract]:Collision detection has a long history in robot motion planning, computer graphics and other fields. In recent years, with the rise of virtual reality, distributed interactive simulation and other technologies, collision detection has become a hot topic. Accurate collision detection plays an important role in improving the authenticity of virtual environment and enhancing the immersion of virtual environment. However, the complexity and real-time of virtual environment have put forward higher requirements for collision detection. Character animation is one of the most active fields in computer science in the last ten years. It has been widely used in many fields such as virtual reality, computer-aided design, advertising media and digital entertainment. With the development of computer hardware technology, especially the development of video card technology with hardware acceleration, real-time character animation has been more and more widely used. Real-time character animation has great economic prospects and theoretical significance. In real-time character animation, skinned skeletal animation is the most widely used animation technology. It combines the characteristics of single mesh animation and joint animation. It not only overcomes the shortcoming that single grid animation needs a large amount of memory, but also does not have the problem of body cracks of 3D characters which often occur in joint animation. In this paper, the collision detection technology of skinned skeletal animation is studied and discussed in detail. In the previous research, there is no special research on the skinned skeletal animation, but only the traditional collision detection technology is applied to the skinned skeletal animation, which makes the pertinence not strong. Not very good for skinny skeletal animation system. Based on the traditional collision detection technology and the characteristics of the skinned skeleton, this paper improves the collision detection algorithm to adapt to the skinned skeletal animation system, and verifies the validity and applicability of the algorithm through experiments. The main work of this paper is as follows: 1 according to the characteristics of skeletal structure in skinned skeletal animation, OBB is chosen as the bounding box of virtual characters, which makes the bounding box and the bone relate to each other. Be a one-to-one relationship. (2) the algorithm of generating parent OBB by two child OBB is improved, and the hierarchical tree generation of role bounding box is improved. Using the characteristics of Animation structure in skinned skeletal animation, the update of OBB hierarchy tree is improved, so that the update of OBB can make use of the information of bone transformation of virtual characters, that is, the change of OBB and the transformation of bone. Speed up the update of OBB trees. Finally, experiments show that the algorithm improves the efficiency of collision detection and meets the needs of collision detection of complex 3D characters in virtual environment.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2008
【分类号】：TP391.41

【共引文献】