STModelViz: A 3D spatiotemporal GIS using a constraint-based approach

Abstract

Advances in data acquisition techniques and model simulations produce increasing volume of 3D spatiotemporal data. However, existing systems provide limited capabilities to manage such data. This paper reports an effort to design and implement a prototype system for 3D spatiotemporal data. Due to the complexity of such data, the ability to verify their integrity is central to the data management system. We adopted a constraint-based approach which addresses data integrity explicitly. In the article, we define constraint conditions, formulate constraints using a formal language we have extended and evaluate constraints using enhanced computational algorithms. We focus on a set of relational integrity constraints pertaining to the spatial, temporal and spatiotemporal properties of 3D spatiotemporal data. We extended the Object Constraint Language (OCL) to handle spatiotemporal (ST) objects. ST–OCL is used to describe and record constraints. Constraints expressed by ST–OCL statements are evaluated by the enhanced algorithms to identify different topological relations between 3D spatiotemporal objects. The prototype system demonstrates how a constraint-based approach can be used to develop DBMS capabilities in managing 3D spatiotemporal objects. Using the dynamic repartitioning of airspace sectors as an application example, we show that the capabilities of the prototype to manage 3D spatiotemporal objects can be customized for specific domain applications.