PutMode: prediction of uncertain trajectories in moving objects databases
Applied Intelligence
| Hi-index | 0.00 |
Moving object databases (spatio-temporal databases) are one of the recent evolutions that emerged to fulfill some of the urging requirements of new applications including: mobile phone services, emergency applications E-911, navigational and military services, flight management and tracking, m-commerce, and various location based services (LBS). Moving object databases integrate traditional spatial and temporal databases and studies modeling, managing and querying objects whose location and/or extent change over time. Modeling and querying moving objects are essential components in moving object databases. In this thesis we focus on designing efficient query languages for moving objects. We study both, objects with exact trajectory information and objects whose location information is not available at every time instant and hence have uncertainty inherent in their location information. Our main contributions are: For exact moving object trajectories (1) we study temporal properties of queries using constraints approach, (2) we propose a distance-based query language for trajectory queries, (3) we propose a query language for reasoning about trajectory properties (i.e. time varying topological properties). For uncertain moving object trajectories (1) we employ techniques from probability theory and model an uncertain moving object trajectory as a stochastic process that follows a probability distribution, (2) we present an evaluation approach for probabilistic range queries, (3) we study the evaluation of a more general class of queries on trajectories with uncertainty.