2可变形网格框架-中国图象图形学报.DOC

中图法分类号：TP391.9 文献标识码：A 文章编号：1006-8961(2016) - 论文引用格式： 可变形网格引导的群体队形仿真 李祖宁12, 何武12 1.四川师范大学可视化计算与虚拟现实四川省重点实验室,成都,610068; 2.四川师范大学影视与传媒学院,成都,610068 摘 要：目的：许多群体运动仿真算法侧重于模拟由大量自由移动个体所组成的群体行为，而针对具有特定队形的群体运动仿真算法较少。为解决这一问题，采用改进的网格引导方法，利用可变形网格对群体运动进行控制。方法：首先，对群体队形进行三角划分，建立一个连接所有智能体的队形网格。然后，利用障碍势场法在群体运动的过程中对队形网格进行变形，使智能体在避免与障碍物发生穿透的同时尽可能保持整体队形稳定。最后，针对障碍物穿过队形网格时可能造成的局部智能体 “错位现象”，提出了基于吸引点的网格引导方法，使群体绕过障碍物后能迅速恢复原来队形。结果：使用Unity软件对军队行进、动态车流、群体表演等场景中的智能体编队移动进行仿真，并设置了不同规模的群组交换实验进行对比。结果表明本文算法的实时计算花销主要集中在网格变形阶段：在队形网格顶点总数为1000时，单位仿真步内网格变形阶段的平均运行时间为20.15ms。队形网格划分阶段是一个预先进行的过程，不影响算法实时性。基于吸引点的网格引导方法提高了智能体的全局移动效率，使队形变化更加自然流畅。结论：可变形网格引导的群体队形仿真算法在群体运动过程中能有效维持队形稳定，无论障碍环境是静态的还是动态的都能实现良好的群体避障，说明算法的有效性。 关键词 ：群体仿真；变形网格；碰撞避免；吸引点；引导网格；障碍势场 Deformable guiding mesh-based simulation of group formation Zuning Li12， He12 1. Visual Computing and Virtual Reality Key Laboratory of Sichuan Province, Sichuan Normal University, Chengdu, 610068; 2. College of Movie and Media, Sichuan Normal University, Chengdu, 610068 Abstract: Objective: Crowd simulation research has become as an increasingly popular topic due to its potential applications in virtual reality and computer animation. Most of the existing researches utilize爐he imethods, such as social force models, hydrodynamic models and data-driven models, to control the movement of the groups. The social force models treat agents as independent individuals with mass and velocity and control the movement of the agents by applying external controlling?force. The hydrodynamic models introduce the concept of fluid?dynamics into crowd simulation, which is?appropriate?for simulating large scale groups. The data-driven models extract the data from the videos of real crowds, and enter the data into the crowd simulation to obtain authentic group behavior. These methods focus on the simulation of the groups which contain a mass of free-move agents. However, they can’t be applied to simulating the groups that move in a specific formation, which extensively exist in