网络空间移动对象模型的应用与发展

海量移动对象轨迹数据的涌现促使移动对象数据库成为一个新兴的研究领域。网络空间是人与交通工具移动的主要空间范围,也是移动对象数据库领域最重要的研究空间。网络空间下的移动对象建模是交通科学、社会与城市计算的基础性关键技术。由于受到所依附的地理空间的限制,网络环境下的移动对象数据表达与自由空间相比具有很大的特殊性,面临新的挑战。本文分析了网络空间移动对象建模的基本需求,对20年来,业界提出的各种网络空间移动对象数据模型进行了系统回顾和分类,深入分析了各种网络空间移动对象数据模型的特征、优点与局限性,提出了现阶段网络空间移动对象数据建模需要关注的关键问题,并对未来研究方向进行了展望。

Advance in Moving Object Data Modeling under Geographic Network Environment

In recent years, along with the rapid development of location technologies such as GPS, RFID and wireless sensor networks as well as the widespread use of location-aware devices such as mobile phones and GPS receivers, large amounts of trajectories of moving objects can be easily acquired. By using moving objects databases, massive trajectories of moving objects can be manipulated and handled for various applications, which makes the study of moving objects databases more and more important. Among the most recent researches, the study space where moving objects travel can be mainly divided into two types, the Euclidean space and the geographic road networks. In Euclidean space, moving objects can move freely. However, under geographic network environment, moving objects are limited to geographic networks and must follow the road regulation. For the latter research branch is much more practical than the former one, this paper concerns the latter branch and focuses on modeling moving objects in geographic networks, which is the foundation of the study of moving objects databases. It has been a hot research topic in the field of moving objects database management and also provides key technology for many other areas such as transportation, location-based services, urban mobility and social computing. Although it is of great theoretical significance and application value, modeling moving object in geographic networks challenges the research community a lot due to network constraints. In this paper, firstly, most existent geographic network-constrained moving object data models over recent decades has been systematically reviewed and classified. The related literatures show that most geographic network-constrained models can be divided into four categories including edge-based network-constrained model, route-based network-constrained model, partition-based network-constrained model and spatial-temporal network-constrained model. Then model characteristics, advantages as well as limitations have been elaborately analyzed, based on which, finally some crucial points on modeling moving objects in geographic networks has been proposed and discussed.