Modeling Historical Trajectory Data Based on Multi-Sensors Environment

摘要

Moving Object Databases should be not only capable of storing and managing spatio-temporal trajectory data effectively, but also support to query spatio-temporal attributes and relations of the trajectories. The key research issue with Moving Object Databases is the representing ofspatio-temporal data. Although the existing spatio-temporal trajectory models have the ability to represent the general trajectory of moving object, they are unable to deal with a complex situation that the moving objects are observed by multi-sensors at the same time, which they may haveseveral similar trajectories and complex relations. In order to solve this problem, we propose a new trajectory data model called Multi-Sensor Historical Trajectory model. In this model, the trajectory is not regard as an attribute of the moving object, but a kind of data objects generatedby both of the moving objects and the sensors. Meanwhile, we also consider the trajectory uncertainty problems in our model and design the related operations. Finally, we show some examples of querying spatio-temporal attributes and the complex relations of trajectories using the proposedmodel and the operations.