虚拟教育环境中虚拟人（角色）技术的研究与应用

【作者】 叶绿；

【导师】 陈纯；

【作者基本信息】 浙江大学 ， 计算机应用技术， 2005， 博士

【摘要】 虚拟教育环境是虚拟现实在教育及培训领域中的应用。虚拟现实在教育中应用的意义在于:高度激发学习动机、突破时间和空间限制、更深入的心理感受、提供学习者尝试不同的选择。虚拟人(角色)是虚拟教育环境的重要组成部分,对于增强场景的真实感、沉浸感具有非常重要的作用。情绪是虚拟人(角色)与虚拟环境相互作用的结果,具有情绪的虚拟人(角色)更能增强虚拟人(角色)的真实性。一般而言,虚拟人(角色)可以视为一个自主的智能主体,具有感觉、知觉、情绪和行为。本文首先讨论了虚拟人(角色)的结构,包括虚拟人(角色)的感知模块、思维模块、行为模块和情绪转换决策模块,这些模块集合成为一个整体。感知模块接收到外面世界的信息后,思维模块会根据虚拟人(角色)的目标和当时的情绪状态来过滤一些不太重要的信息,同时把那些重要信息传送给情绪转换决策模块,由情绪转换决策模块根据一定的规则来决定虚拟人(角色)情绪的变化。虚拟人(角色)的思维模块能够根据信息库中定义的规则来决定外部被感知到的信息是否产生新的情绪增量矩阵。本文提出了一个新的情绪转换决策模型,该模型是基于OCC情绪模型之上,并能根据虚拟人(角色)的性格特点来确定一个个性矩阵。根据个性对情绪影响的大小,可以确定一个情绪个性矩阵(MPE)。思维模块在接收到有效信息后,把他们转化成情绪增量矩阵,由上面三个矩阵最终来决定虚拟人(角色)情绪变化。另外,本文还提出了一个新的实现虚拟人(角色)思维模块的方法,其中的思维模块是用Java中的JESS工具来完成的。为使人们能在计算机环境中模拟实际或虚拟的声场获得与临场相似的听觉感受,本文介绍了一个虚拟声环境构造系统主要实现方法,阐述了该系统主要的实现技术。利用与头相关的传递函数实现人耳的滤波效应。在模拟静止声源的基础上,系统地实现了运动声源的模拟。本文还实现了E-Teatrix系统和虚拟人(角色)动作库AgentLib。E-Teatrix是利用虚拟人(角色)技术来辅助中小学生来学习的一个软件,小朋友们可以根据读到的故事更多还原

【Abstract】 Virtual educational environment is the application. The importance of applying virtual reality lies in the following aspects: inspire the motivation of learning, free of the limitation of space and time, immersive feeling and options of the learners.Virtual synthetic character is a key component of virtual environment, which can increase the reality and immersive feeling of the scene. Emotion is the interactive result of the virtual character and virtual environments. A virtual character has emotions which can greatly enhance the realistic of the scene. In general, a virtual synthetic character can be regarded as an autonomous agent with sense, perception, emotion, behavior and action.The architecture of intelligent virtual synthetic character is discussed in this paper. A virtual character is composed of many modules, including sense module, perception module, plan module, behavior module and the decision-making module of the emotion. These modules are integrated into a system. The sense model receives external stimulus information. After filtered the unimportant information, the mind model sends this information to decision-making model of the emotion. The decision-making model of the emotion decides which sort of emotion should be selected according to defined rules.According to the rules in the knowledge base and the information sensed by sensors, the mind will determine whether the new incremental matrix should be produced.The new concept of decision-making model of the emotion is also introduced. This model is based on the OCC emotion model. According to the specialty of the virtual character, the decision-making model of the emotion deduces a personality matrix and emotional personality matrix. Having received the valid information, the mind then turns information into incremental matrix. These three matrixes determine the emotion changes of virtual character. This paper also presents a way to implement the mind of virtual synthetic character. The mind module is implemented by JESS (Java Expert System Shell).For people having the similar hearing feeling with real sound field in computerenvironment where real or virtual sound field is simulated, the research of virtual acoustic environment is a cannot missing part of the research of interactive virtual environment. This paper introduces the primary realization methods and techniques of a system to construct virtual acoustic environment.Based on the study on the algorithms and techniques of virtual human and behavioring modeling, the author implemented two prototype systems (E-Teatrix, and AgentLib). E-Teatrix is a system which uses virtual character to aid students learning. Using 3D virtual characters and virtual environment is the most advantage of E-Teatrix. AgentLib is a 3D virtual character motion library, the aim of the library is to provide a 3D rendering engine for upper level AI usage.更多还原