普适环境中基于情境时序关系的活动监控研究

发布时间：2018-06-23 09:06

本文选题：普适计算 + 情境　；参考：《大连理工大学》2013年硕士论文

【摘要】：普适计算是继桌面计算模式后的新兴计算模式,它强调物理空间与信息空间的融合。随着普适计算的发展,利用情境感知技术为处于工作状态下的用户提供透明的计算服务己成为普适计算领域的研究问题之一。研究表明对于那些需要通过多个规范化操作完成的活动,往往由于活动执行条件的不满足、用户异常行为以及实际操作中的不当(如遗漏操作、操作顺序错误等),而无法实现活动目标,甚或引发一些安全事故。为了避免上述情形的发生,就需要对活动执行过程进行实时地监控,并根据监控结果,给用户提供适合于当前活动状态和地点的辅助服务,以达到规范用户操作的目的。鉴于以上分析,本文针对普适计算的研究背景,对如何利用与活动有关的情境信息实现活动监控进行研究。本文中考虑的活动是在一定的客观条件下,经由若干个动作或规范化操作完成,其执行过程中蕴含一定的时序信息,而这些时序信息也为此类活动的监控提供了证据与线索。因此,为了根据嵌入在物理空间中传感器等计算设备感知到的情境来监控活动中的异常情况,这就需要准确地描述活动中的时序信息。本文首先结合所研究的问题,阐述了基于情境感知技术的活动监控系统框架。考虑到活动执行过程中,情境是动态变化的,给出了情境模式的概念,并定义了情境模式之间的时序关系,而通过情境模式之间的时序关系能够清晰地表达活动中的时序信息。根据活动正常执行时各情境模式之间的时序关系,归纳了活动实际执行过程中的异常模式,这些异常模式的表现形式主要为情境模式之间不满足对应的时序关系,并给出了判断异常模式的方法。最后,本文以金相组织分析为研究案例,构建了基于情境时序关系的活动监控模型,并针对案例中的镶嵌操作,根据操作过程中不同时刻的情境取值,来判断其中的异常情况,验证了方法的可行性。
[Abstract]:Pervasive computing is a new computing model after desktop computing. It emphasizes the integration of physical space and information space. With the development of pervasive computing, the use of situational awareness technology to provide transparent computing services for users at work has become one of the research problems in the field of pervasive computing. Studies have shown that for activities that need to be performed through more than one normalized operation, user behavior is often abnormal due to unsatisfactory execution conditions of the activity, as well as missteps in actual operations, such as missing operations, Incorrect sequence of operations, etc., without achieving the active goal, or even causing some safety incidents. In order to avoid the above situation, it is necessary to monitor the process of activity execution in real time, and according to the monitoring results, to provide users with auxiliary services suitable for the current state and location of activities, in order to achieve the purpose of standardizing the operation of users. In view of the above analysis, this paper studies how to use situational information related to activities to realize activity monitoring in view of the research background of pervasive computing. The activities considered in this paper are completed by a number of actions or standardized operations under certain objective conditions, which contain certain temporal information in the execution process, and these temporal information also provide evidence and clues for the monitoring of such activities. Therefore, in order to monitor the abnormal situation of the activity according to the situation perceived by the sensor and other computing devices embedded in the physical space, it is necessary to accurately describe the temporal information in the activity. In this paper, the framework of activity monitoring system based on situational awareness technology is introduced. Considering that the situation is dynamic in the process of activity execution, the concept of situational pattern is given, and the temporal relationship between situational patterns is defined. The temporal information of activity can be clearly expressed by the temporal relationship between situational patterns. According to the temporal relationship of each situation pattern when the activity is executing normally, the abnormal patterns in the actual execution process of the activity are summarized. The main manifestation of these abnormal patterns is that the situation pattern does not satisfy the corresponding temporal relation. The method to judge the abnormal pattern is also given. Finally, taking metallographic organization analysis as a case study, this paper constructs an activity monitoring model based on situational temporal relationship, and according to the mosaic operation in the case, according to the situation values at different times in the operation process, to judge the abnormal situation. The feasibility of the method is verified.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP338;TP277

【参考文献】