基于完全二叉树BVH的自碰撞检测算法

计算机图形学与虚拟现实应用的一个主要任务是对柔性物体的动态实时模拟。这类仿真任务需要能够快速计算的模型与动力学仿真算法,特别是高效的碰撞、自碰撞检测算法。尽管研究者们针对刚体的碰撞检测已经做了大量工作,但柔性物体对碰撞检测提出了更高的挑战。给出一种基于完全二叉树的包围盒层次结构,并运用基于此结构的构建、更新算法,优化仿真系统中数个环节,实现复杂场景中布料类柔性物体的自碰撞检测与物理模拟。     

快速的布料碰撞检测和有效的接触摩擦算法

为了快速处理布料的碰撞检测并获得真实的接触摩擦仿真效果,提出一种基于罚函数的碰撞/接触解决方案.首先,采用质点-弹簧模型进行布料的仿真模拟,在弹簧形变方向添加改进的阻尼力,以减少粒子之间的振荡来保证系统稳定性;其次,采用代数非穿透滤波器对连续碰撞检测算法进行简化求解,快速判断是否存在方程根,提高布料每帧运行的仿真效率;最后,采用库仑约束和接触约束对每个碰撞/接触对进行约束,并结合改进的罚函数法有效地响应所有的碰撞/接触对.实验结果表明,该算法在CPU仿真环境下能快速有效地处理布料的碰撞和接触摩擦,模拟出布料复杂的物理行为,适用于实时的交互应用.     

基于OBB包围盒的碰撞检测研究与应用

碰撞检测是计算机仿真、虚拟现实等领域内的研究重点之一,其中基于包围盒的碰撞检测算法是一种比较方便有效的方法。为了达到仿真系统中所需要的碰撞检测的精确性和实时性,对方向包围盒(OBB)的计算方法进行了分析,并针对其不足之处进行了改进和优化。利用改进后的OBB包围盒碰撞检测技术实现了JQ900型下导梁架桥机中部件间碰撞的精确快速检测,并最终在VC++和Open Inventor API平台上实现了系统的逼真的仿真效果。     

基于Voronoi图的实时刚体破碎模拟

为了实现具有较强物理真实感的刚体破碎实时模拟,提出了一种基于预计算和Voronoi图的刚体破碎模拟方法.首先分析脆性刚体材料破碎过程,得到拟合性较好的正态分布种子点,在此基础上采用脱机增量算法和格雷厄姆凸包生成算法,生成刚体破碎的Voronoi多边形;然后根据能量守恒方程和材料力学分析,计算破碎区和裂纹区的半径,并根据碰撞过程中的能量转换约束条件推导出破碎后的碎片速度;最后计算碎片间的碰撞检测以及碰撞后的速度、姿态等,得到碎片的运动状态.实验结果表明,该方法不仅可以满足游戏、仿真中交互的实时性需求,同时碎片的产生以及运动符合物理规律,具有较强的真实感.     

凸多面体连续碰撞检测的运动轨迹分离轴算法

针对凸体间的连续碰撞检测,在距离算法(Gilbert-Johnson-Keerthi distance algorithm,GJK)基础上,提出一种采用运动轨迹分离轴计算的线性连续碰撞检测算法.该算法首先采用支撑点和投影技术,剔除必定不发生碰撞的物体,以加速碰撞检测的速度;然后,对可能发生碰撞的物体,计算2个凸体的Minkowski差集,所形成的凸包与运动路径执行GJK分离轴算法,实现在整个时间区间内一次性完成碰撞检测任务;最后,采用几何方法以及超平面与射线求解方式计算射线与凸体边界近交点,确定出第一次发生碰撞位置,并调整运动物体位置,完成碰撞响应过程.该算法不需要构造扫掠体,连续检测过程中不需要凸体间的求交计算.将文中算法应用于物体方向包围盒的连续碰撞检测,算法分析和实验结果表明,该算法对包围盒的连续碰撞检测具有较高检测精度和响应速度.     

三维单目虚实物体球对面碰撞检测算法

为加强增强现实的沉浸感与真实性,实时地进行虚实物体间的碰撞检测至关重要.因此,提出一种基于增强现实和单目视觉的任意形状虚实物体碰撞检测估计算法.通过改进现有的单目二维虚实碰撞检测及响应算法,针对现有碰撞检测算法存在的计算复杂度高的问题,提出一种仅需计算实际物体4个特征点的三维碰撞检测算法;并通过对象分割、特征点提取、碰撞检测和碰撞响应等过程取得与真实世界物理特性一致的三维虚实碰撞响应估计效果.在增强现实有标记和无标记环境下分别进行实验结果表明,碰撞检测的计算量与二维算法相近,且具有景深效果,实现了基于单目视觉的增强现实三维虚实碰撞响应预测和处理.     

虚拟手术中基于可碰撞集的软组织自碰撞检测算法

快速精确的碰撞检测是保证手术仿真真实性的重要条件.由于虚拟手术仿真过程中的高复杂性和高计算性,使得现有的碰撞检测已经不能满足实时性与复杂性的要求,为此提出一种新的软组织自碰撞检测方法.该方法根据软组织的曲率提取可碰撞集,基本几何元素间的自碰撞仅在该可碰撞集中发生,基本几何元素相交检测采用基于特征检测的方法.通过试验表明,该方法不仅能有效地提高碰撞检测的速度,而且能对人体软组织进行相对精确的检测.     

基于改进轴向包围盒的碰撞检测算法

研究机场应急救援是民航领域的问题.为防止机场地面车辆拥堵,要求在虚拟场景下的应急演练可全方位测试系统.为增强虚拟演练场景中碰撞检测的真实性与精确性,提出了一种基于特征三角形的碰撞检测算法.在基于三角形的模型中,通过在三角形中添加特征元素(点、边、面)形成特征三角形,利用特征三角形可以有效解决重复查询;通过轴向包围盒结合特征三角形,更好地完成碰撞检测,最后进行精确求交计算,为碰撞响应提供更多的碰撞信息.实验结果表明,算法可以缩短计算时间,提高检测精度,具有实际指导价值.     

基于两级非渗透滤波的连续碰撞检测算法*

为提高高精度模型在物理仿真中的碰撞检测效率,提出了一种基于两级非渗透滤波的连续碰撞检测算法,对无碰撞可能性的基本图元对进行快速剔除。首先,通过给基本图元增加额外的层次包围盒,对没有发生重叠的基本图元对进行第一级滤除。其次,采用代数滤波器对通过第一级滤除的基本图元对进行低层滤除。实验结果表明,本文的方法不仅可以有效地检测出碰撞信息,而且可以在物体发生大幅度形变时发挥较好的滤波特性。     

虚拟手术实时物体碰撞检测和软组织变形研究

针对虚拟手术系统中实时碰撞检测和软组织变形的问题进行了研究。对于实时碰撞检测,使用了包围盒检测结合三角面片检测的方法来检测物体之间的碰撞,并且找到相互碰撞的两个三角面片。对于软组织变形,使用了一种基于胡克定律的弹簧质点模型来虚拟仿真变形,并提供了控制点移动的方法。采用优化虚拟手术中的物体碰撞检测方法,减少了不必要的计算和增强了系统的精确度;简化了软组织变形的算法,相对加快了有限元模型运算速度。对推动虚拟手术的发展具有积极的意义和作用。     

流体模拟动画技术新进展

自然场景的模拟一直是计算机图形学研究的热点,随着动画影视、虚拟现实以及计算机游戏的飞速发展,对流体模拟技术的绘制速度与场景细节真实感的要求也在不断提高。通过描述最近几年来流体动画模拟领域的新技术,对流体模拟计算仿真的主要步骤细节技术进展进行分析讨论,并指出流体动画模拟技术进一步的发展方向。     

Research problems in clothing simulation

Clothing simulation and animation are of great importance in computer animation. If cloth simulations could be improved to the point that they could generate realistic cloth motion in real-time, they would find uses in many aspects of daily life such as in fashion design and manufacturing. The area of cloth simulation and animation is full of technical challenges: creating more realistic results, achieving faster run-times, and developing methods capable of constructing and simulating more complex garments. This paper provides an overview of the key procedures involved in the creation of clothed characters, describes the current state-of-the-art techniques, and proposes the research problems that most require further study. Three technical aspects of cloth simulation are considered in this paper: garment construction, physically based simulation, and collision resolution.     

A collision resolution algorithm for clump-free fast moving cloth

Cloth animation is an important area of computer graphics due to its numerous applications. However, so far a fast moving cloth with multiple wrinkles has been difficult to animate because of the cloth clump problem. Cloth clumps are the frozen areas where cloth pieces are clustered unnaturally — an obstacle in making a realistic cloth animation. Hence we present a novel cloth collision resolution algorithm that prevents clump formation during fast cloth motions. The goal of our resolution algorithm is to make cloth move swiftly without having any unnatural frozen cloth clumps, while preventing any cloth-cloth and rigid-cloth penetrations at any moment of a simulation. The non-penetration status of cloth is maintained without the formation of cloth clumps regardless of the speed of cloth motion. Our algorithm is based on a particular order that we found in the resolution of cloth collisions, and can be used with any structural modeling approaches such as spring-masses or finite elements. This paper includes several realistic simulation examples involving fast motions that are clump-free.     

Laying out objects with geometric and physical constraints

Modeling scenes involves two tasks:object modeling andobject layout. This paper focuses on object layout and proposes a constraint-based approach which yields a powerful object layout environment. The approach uses collision detection and physical simulation to ensure geometric and physical consistency of the resulting scenes, such as no interpenetration, and physical stability of the objects. A prototype system is developed, providing six basic operations; PUT, PUSH/PULL, TURN/TILT, PICK-UP, TRANSLATE, and ROTATE. The system: ensures geometric and physical consistency; provides easy-to-use operations analogous to object placement in real life; allows twodimensional control easily specified by mouse. Interactive speed is achieved on graphics workstations by using rasterized collision detection and simple quasi-static motion simulation. The system is interfaced to modeling/rendering/animation systems, and realizes an integrated environment for object modeling, object layout, rendering, and animation. We describe several scenes that have been modeled using the system and argue that these experiments confirm that the scene modeling task is greatly simplified by our constraint-based approach.     

基于虚拟现实的三维园林景观设计仿真系统

现有的园林景观专业CAD软件对三维模型及虚拟仿真效果表达有限,只能进行二维施工图绘制.为解决动态景观特性,研发了虚拟现实技术的三维园林景观设计仿真系统,采用面向对象的参数化三维实体建模手段,实现了三维实体建模、设计场景的三维实时虚拟呈现、渲染图及动画制作、施工图绘制等功能,并探索了系统计算原理、实现方法及OpenGL扩展功能等相关技术.然后简要介绍了系统整体功能,详细论述了专业模块中的仿真过程.最后以工程实例验证了系统的有效性.     

利用OSG三维图形渲染引擎实现的骨骼动画

动态实体模型是虚拟场景的重要组成部分,骨骼动画是表达虚拟场景中动态实体行为的主要技术。提出了一种虚拟场景中三维模型骨骼动画渲染的策略,即采用第三方软件建立三维动画模型,基于Cal3D动画控制机制,使用OSG三维图形渲染引擎对虚拟场景中运动的三维模型进行模拟。在VisForest软件的基础上,集成了骨骼动画渲染方法,解决了动物在虚拟地表上“踏空”的问题,扩展了动态实体在虚拟环境中的行为模拟。以梅花鹿和大象为例,逼真地模拟了动物在虚拟环境中的奔跑和行走的过程,模拟过程中动作衔接比较自然。     

布料仿真建模研究进展

布料仿真一直是计算机动画中的研究热点与难点,对提高计算机动画质量以及用户体验具有重要意义,布料是一种非常经典的柔性材料物体,遍布于人们的日常生活中。虚拟世界中虚拟角色强烈的视觉真实感主要来源于逼真的虚拟人物的服装动画,这在很大程度上可以增强用户的体验感,在游戏娱乐、电影电视和动画制作等领域有着十分广泛的应用前景。布料仿真的质量与速度直接决定了计算机动画的整体水平,而布料的模拟水平则起着至关重要的作用。随着计算机软硬件的不断发展和计算机动画市场需求的提高,对布料仿真建模方法的研究受到越来越多的关注,布料仿真建模方法也因此有了较大发展。本文通过回顾布料仿真建模方法的相关工作,对国内外方法的研究进展进行综述,从布料仿真中数值积分方法的改进、多分辨率网格的改进和机器学习方法的使用等方面对布料仿真方法的发展进行简要阐述,并根据不同方法在布料模拟应用上的特性,对几大类改进方法进行了相应的总结与展望。同时选取几种算法进行对比,并给出建议。     

The HUMANOID Environment for Interactive Animation of Multiple Deformable Human Characters

We describe the HUMANOID environment dedicated to human modeling and animation for general multimedia, VR, and CAD applications integrating virtual humans. We present the design of the system and the integration of the various features: generic modeling of a large class of entities with the BODY data structure, realistic skin deformation for body and hands, facial animation, collision detection, integrated motion control and parallelization of computation intensive tasks.     

Markov-Type Vector Field for endless surface animation of water stream

In this paper, we propose a hybrid (physical-stochastic) model of surface element (surfel) fluctuations for the visual simulation of an endlessly running water surface. This model comprises two main phases: preprocessing and endless animation phases. First, we simulate a physics-based method for a specific period of time during the preprocessing phase. We construct a stochastic vector field in the simulation, referred to as a Markov-Type Vector Field (MTVF), using only the surface values of the fluid flow. Next, we import the MTVF data into the main endless animation phase and create a surface fluctuation animation by surfels and temporary velocity field modeling of the MTVF using a random sample. In our approach, the surfel edges that cover the fluid flow domain are transferred simply via a temporary single velocity and the new flow surface is determined directly based on their positions. MTVF allows us to generate a water surface animation endlessly in real time without the time-consuming processes of solving the corresponding physical equations. We describe the MTVF construction method and the endless surface animation steps, as well as present the results of experiments that demonstrate the plausibility of our method.     

A survey of CAD model simplification techniques for physics-based simulation applications

Automated CAD model simplification plays an important role in effectively utilizing physics-based simulation during the product realization process. Currently a rich body of literature exists that describe many successful techniques for fully-automatic or semi-automatic simplification of CAD models for a wide variety of applications. The purpose of this paper is to compile a list of the techniques that are relevant for physics-based simulations problems and to characterize them based on their attributes. We have classified them into the following four categories: techniques based on surface entity based operators, volume entity based operators, explicit feature based operators, and dimension reduction operators. This paper also presents the necessary background information in the CAD model representation to assist the new readers. We conclude the paper by outlining open research directions in this field.