Querying Moving Objects in GeoPQL

In recent years, spatio-temporal databases are receiving considerable attention due to the extensive amount of applications dealing with objects that change their spatial characteristics over time. For example, applications for tracking vehicle (cars, trains, aircrafts or ships), environmental change monitoring or the analysis of group/individual movement behaviour require the storage and querying of spatial objects changing continuously over time (moving objects). To consider together spatial and temporal features of data is the first challenging requirement of spatio-temporal database and consequently it represents also the topic of considerable research activity. Another important research topic concerns the development of friendly and simple query languages that allow to ask for temporal changes in spatial situations. Although traditional textual query languages have been proven to enable the user to express complex requests, the ease of use of such languages is very limited, particularly for non-expert users. Thereby, the visual-based querying paradigm has emerged as the most intuitive and natural one, trying to offer a good trade-off between expressiveness and intuitiveness. In particular, a large number of visual languages has been proposed to query spatial database, as spatial relationships are well suited to be represented through visual notations. Temporal changes of spatial relationships are more difficult to be visually expressed due to the immaterial nature of time. Therefore, the graphical representation of the temporal dimension is a challenge for visual spatio-temporal query languages. The aim of this paper is to gather the two challenges described above by proposing an extension of the Geographical Pictorial Query language (GeoPQL) [Ferri et al. 2005] in order to support visual spatio-temporal queries. The main contribution is the representation of queries on moving objects in a three-dimensional space where the spatial dimension is described by a two-dimensional layer and the temporal dimension is the third dimension of this space and, therefore, the overlap of layers allows to represent the temporal changes of spatial situations. In this paper, after a brief discussion about existing research activities related to this work, we analyze the main characteristics of GeoPQL focusing on the spatio-temporal operators and their visual representation.