【作者】 姜波；

【导师】 陈纯；

【作者基本信息】 浙江大学 ， 计算机软件与理论， 2007， 博士

【摘要】 计算机支持的协同编辑支持在地理位置上分散的用户通过网络同时浏览和编辑文本/图形/图像/多媒体文档,其研究成果可以广泛地应用于协同编著、协同产品设计、协同图案设计以及视频会议等多个领域。协同感知是协同编辑重要的支撑技术,良好的协同感知是提升协同能力和工作效率的关键。随着移动计算和嵌入式计算技术的飞速发展,普适计算使得构筑群体之间更为便捷、灵活、和谐的普适协同工作环境成为可能。然而,普适环境所具有的网络连接及传输的不稳定性、多种计算设备资源的不均衡性,使得普适环境中协同感知的研究面临众多新的挑战,相关研究具有重要的理论与现实意义。本文结合国家自然科学基金项目(60573176)“面向普适环境的协同编辑中若干基础问题的研究”和浙江省自然科学基金重点项目(Z603231)“基于互联网的图案协同CAD/CAM关键技术研究”的研究内容,基于认知心理学的相关理论,提出了融合人类认知特征的普适环境下协同感知研究思路,建立了相关的模型、方法和算法,同时通过CoDesign原型系统进行了实现和验证。本文的主要研究内容如下:首先,对协同感知相关的研究现状进行分析,提出了普适环境下协同感知的特点以及面临的问题与挑战,结合认知心理学中人类认知特征,提出了协同感知的若干研究问题。然后,针对普适环境下协同终端资源不均衡性和认知资源有限性,提出了非对称协同感知映射模型和机制,以及相关的共享工作空间逻辑一致性维持方法。针对人类选择性注意特点,依据认知心理学注意资源理论,提出了注意驱动的感知关注区预测及算法,使得在非对称映射中映射区域为协作者最可能关注的感知内容,减少了协同感知所需资源。接下来,针对普适环境下网络延时、丢包和抖动对远程动态操作过程带来的间歇性呈现问题,依据人类认知对感知信息的完整性要求,提出了基于AFEC算法缓解网络数据包丢失引起的操作控制点信息传输不准确问题。提出了远程控制点运动轨迹预测Scale-Changeable Machine-Learning算法,使得远程动态感知信息轨迹的显示平滑而自然。最后,针对普适环境下协同编辑意图感知困难及分布式认知中合适的表征形式有助于减少认知难度的特点,提出了基于图形、图像等多编辑实体对象的语义感知模型、语义维持方法、语义冲突检测与解决算法,提供了更高层次感知支撑技术。更多还原

【Abstract】 Computer supported collaborative editing systems allow a group of users to view and edit the same text/graphic/image/multimedia document simultaneously from geographically distributed sites connected by networks. They have many applications such as collaborative authorizing, collaborative product designing, collaborative pattern designing, video netmeeting and etc. Collaborative awareness, as an important technology for collaborative editing, will promote the collaborative capabilities and enhance the efficiency of group work.With the development of mobile computing and embedded computing, ubiquitous computing makes it possible for group to create the flexible and harmonious collaborative working environment. However, in ubiquitous environment the network connection and transmission are unstable and the resources of computing devices are quite different, which brings great challenges to the research of collaborative awareness. Therefore, it is important to research the related techniques.The investigation is combined with key issues in the research project of "Research on basic problems of collaborative editing in ubiquitous environment" and "Research on key technologies in the Internet-based collaborative pattern CAD/CAM system". Based on thories of cognition psychology, we research on collaborative awareness based on the characteristics of cognition. Related models, methods and algorithms are proposed and realized in our prototype system. Main research work is as follows:First, we analyze current research topics and key problems of collaborative awarenesss, the main characteristics of collaborative awareness in ubiquitous environment and challenges collaborative awareness meat. We present our research issues based on the characteristic of cognition.Second, we analyze the difference of the resources of collaborative sites and the limitation of cognition resources. We present the unsymmetrical awarenss mapping model, mechanism and related algorithm for maintaining the logical consistency of shared workspace. As attention is selective, based on the theory of attention resources in cognition psychology, we propose the algorithm of prediction for regions of interest. The prediction assures that the mapped awareness regions will be interested to users and the resoures that collaborative awareness needs is decreased. Third, network delay, packet loss and jitter brings halting and jumping presence of remote operating and it is necessary for people to gain better cognition based on integrated information. We present the algorithm of AFEC to solve the problem that brought by packet loss. And also, we propose Scale-Changeable Machine-Learning algorithm to predict the trace of the motion of remote control points in order to make the presence smooth.Finally, as appropriate ways of presenting awareness information will promote cognition ability and it is difficult to know user’s editing intention in ubiquitous environment, we present the semantic awareness model based on various types of objects such as graphic and image. Related techniques like sematic maintenance, semantic conflict detection and solution are proposed to provide better awarenss.更多还原