The emerging availability of sensor technologies for measuring properties of the physical space enables everyday artifacts to become aware of their spatial situation – such as where they are located or whether two of them are near or far apart from each other – and adapt to changes accordingly. We specifically addressed the use of spatial relations between networked embedded systems as well as their changes over time. A key aspect is the abstraction of spatial contexts in order to separate details which are not relevant for a certain application, and thereby save computational resources and provide spatial information in a way that is closer to human concepts of space.

For the representation of spatial contexts, we used two types of abstraction:

  • Quantitative abstractions are used for representing the spatial properties of embedded systems. A concept referred to as “Zone-of-Influence”, which represents a geographic region that is of relevance for and has an influence at a particular spatially aware application, has been developed for this purpose. Embedded systems are associated with one or more Zones-of-Influence at a time. An ad-hoc exchange of these Zones-of-Influence enables the recognition of spatial relations, such as distance and orientation relations as well as topological relations between their spatial extensions.
  • Qualitative abstractions such as “left” and “near” are used for representing spatial relations. Compared with quantitative approaches, they are often the preferred choice when the spatial cognition of humans is involved. Thus, it should be easier for a programmer to develop applications by using such high-level abstractions. Moreover, computationally less demanding reasoning techniques can be implemented due to the separation from details that are not relevant for a certain application. We used qualitative spatial relations as the basic building blocks for the development of spatially aware applications.

We have developed a flexible and modular software framework, with the aim to facilitate the implementation of spatially aware applications for mobile and ad-hoc networked embedded systems. It has beenĀ built using the Java-based Equinox OSGi framework as runtime platform. The architecture of the presented framework consists of multiple layers through which low-level spatial context information is processed. The layers are determined by the functional requirements of the framework, which should serve as a middleware between the spatial sensors and the communication module on the one hand, and the application layer for running spatially aware applications on the other hand. These requirements are basically (i) the acquisition and representation of spatial contexts from local sensors or by exchanging Zones-of-Influence with other mobile systems in range, (ii) the application-dependent recognition and qualitative abstraction of spatial relations by comparing local with received Zones-of-Influence as well as (iii) means for drawing conclusions from recognized relations using a rule-based approach.

Images

Videos

The video of a spatially aware application, which notifies the user about surrounding objects by means of vibro-tactile notifications, is based on the above software framework and has been developed in cooperation with Siemens AG Corporate Technology in Germany.


Publications

2010 C. Holzmann, A. Ferscha – A Framework for Utilizing Qualitative Spatial Relations between Networked Embedded Systems – PERVASIVE AND MOBILE COMPUTING, Vol. 6, No. 3, 2010, pp. 362-381. [elsevier] [pdf]

Partners

Researchers

Clemens Holzmann
Department of Mobile Computing, University of Applied Sciences Upper Austria, Austria
clemens.holzmann [at] fh-hagenberg.at

Alois Ferscha, Manfred Hechinger
Institute for Pervasive Computing, Johannes Kepler University Linz, Austria
alois.ferscha [at] jku.at