Wolfgang Gottesheim

Editorial: BeAware!-Situation awareness, the ontology-driven way

Information overload is a severe problem for human operators of large-scale control systems as, for example, encountered in the domain of road traffic management. Operators of such systems are at risk to lack situation awareness, because existing systems focus on the mere presentation of the available information on graphical user interfaces-thus endangering the timely and correct identification, resolution, and prevention of critical situations. In recent years, ontology-based approaches to situation awareness featuring a semantically richer knowledge model have emerged. However, current approaches are either highly domain-specific or have, in case they are domain-independent, shortcomings regarding their reusability. In this paper, we present our experience gained from the development of BeAware!, a framework for ontology-driven information systems aiming at increasing an operators situation awareness. In contrast to existing domain-independent approaches, BeAware!s ontology introduces the concept of spatio-temporal primitive relations between observed real-world objects thereby improving the reusability of the framework. To show its applicability, a prototype of BeAware! has been implemented in the domain of road traffic management. An overview of this prototype and lessons learned for the development of ontology-driven information systems complete our contribution.

On Optimization of Predictions in Ontology-Driven Situation Awareness

Systems supporting situation awareness in large-scale control systems, such as, e.g., encountered in the domain of road traffic management, pursue the vision of allowing human operators prevent critical situations. Recently, approaches have been proposed, which express situations, their constituting objects, and the relations in-between (e.g., road works causing a traffic jam), by means of domain-independent ontologies, allowing automatic prediction of future situations on basis of relation derivation. The resulting vast search space, however, could lead to unacceptable runtime performance and limited expressiveness of predictions. In this paper, we argue that both issues can be remedied by taking inherent characteristics of objects into account. For this, an ontology is proposed together with optimization rules, allowing to exploit such characteristics for optimizing predictions. A case study in the domain of road traffic management reveals that search space can be substantially reduced for many real-world situation evolutions, and thereby demonstrates the applicability of our approach.

A Two-Layered Deployment Scheme for Wireless Sensor Network based Location Tracking

The tracking of objects and humans has recently received a lot of attention as a tool to improve business processes, occupational and public safety. In wireless sensor networks location information is needed to put sensor readings into geographical context. In industrial environments wireless sensor networks can facilitate deployment of tracking applications as they can establish a standalone communication infrastructure (so called mesh networks). Nevertheless the communication infrastructure and thus the tracking applications rely on radio transmission. Therefore problems like ambiguous locations and the neighboring-room problem known from other radio-based tracking solutions may occur. We propose a two layered deployment scheme for wireless sensor networks to overcome these limitations. It consists of a robust communication layer and a flexible location layer, which enable per- room accuracy. Such accuracy is sufficient for many types of tracking applications.

Situation Prediction Nets

Situation awareness in large-scale control systems such as road traffic management aims to predict critical situations on the basis of spatio-temporal relations between real-world objects. Such relations are described by domain-independent calculi, each of them focusing on a certain aspect, for example topology. The fact that these calculi are described independently of the involved objects, isolated from each other, and irrespective of the distances between relations leads to inaccurate and crude predictions. To improve the overall quality of prediction while keeping the modeling effort feasible, we propose a domain-independent approach based on Colored Petri Nets that complements our ontology-driven situation awareness framework BeAware!. These Situation Prediction Nets can be generated automatically and allow increasing (i) prediction precision by exploiting ontological knowledge in terms of object characteristics and interdependencies between relations and (ii) increasing expressiveness by associating multiple distance descriptions with transitions. The applicability of Situation Prediction Nets is demonstrated using real-world traffic data.

Same, Same but Different A Survey on Duplicate Detection Methods for Situation Awareness

Systems supporting situation awareness typically deal with a vast stream of information about a large number of real-world objects anchored in time and space provided by multiple sources. These sources are often characterized by frequent updates, heterogeneous formats and most crucial, identical, incomplete and often even contradictory information. In this respect, duplicate detection methods are of paramount importance allowing to explore whether or not information having, e.g., different origins or different observation times concern one and the same real-world object. Although many such duplicate detection methods have been proposed in literature--each of them having different origins, pursuing different goals and often, by nature, being heavily domain-specific--the unique characteristics of situation awareness and their implications on the methods applicability were not the focus up to now. This paper examines existing duplicate detection methods appearing to be suitable in the area of situation awareness and identifies their strengths and shortcomings. As a prerequisite, based on a motivating case study in the domain of road traffic management, an evaluation framework is suggested, which categorizes the major requirements on duplicate detection methods with regard to situation awareness.

Large-Scale Industrial Positioning and Location Tracking Are We There Yet?

The tracking of objects and humans has recently received a lot of attention as a tool to improve business processes,occupational safety and public safety. Employers and legislators are extremely interested in improving occupational safety as numerous internal executive decrees as well as safety laws suggest. Nevertheless, a recent report on industrial injuries shows that despite remarkable declines over the last ten years further improvement can be achieved with new approaches which are less expensive, more convenient, and truly ubiquitous. The question arises, however, as to whether these available technologies are really feasible in terms of accuracy, usability, cost, and ubiquity. In fact, our analysis of numerous theoretical approaches has revealed several shortcomings in their actual use, especially in harsh industrial environments. In this paper we present the requirements for a largearea industrial location tracking system for improving ccupational safety. We discuss current research and evaluate how well these systems meet crucial requirements.

Making workflows situation aware: an ontology-driven framework for dynamic spatial systems

Business processes face constantly changing context factors like varying customer behavior or market conditions that force to adapt the underlying workflows to these evolving situations. Information overload induced by the diversity of context factors, however, leads to the inability to provide coherently modeled, comprehensible, and re-usable workflows and the failure to recognize relevant situations in time. The main goal of our research project ProFlow is to leverage situation awareness in all phases of workflow management especially focusing on dynamic spatial systems as encountered, e.g., in the domain of road traffic management. ProFlow thereby bases on a generic ontology-driven framework for situation perception and comprehension. This paper details on the corresponding ontological representations especially addressing extension points that allow developers to extend and configure our framework for their own application domains. This forms the basis for the overall system architecture, which is laid out along its prototypical implementation.

Towards duplicate detection for situation awareness based on spatio-temporal relations

Systems supporting situation awareness typically integrate information about a large number of real-world objects anchored in time and space provided by multiple sources. These sources are often characterized by identical, incomplete, and even contradictory information. Because of that, duplicate detection methods are of paramount importance, allowing to explore whether or not information concerns one and the same real-world object. Although many such duplicate detection methods exist, a recent survey revealed that the characteristics of situation awareness--highly dynamic and vague information, which is often available in qualitative form only--are not supported sufficiently well. This paper proposes concepts for qualitative duplicate detection to cope with these key issues of situation awareness based on spatio-temporal relations between objects.

Towards a unified location tracking system for heterogeneous industrial environments

Current research on localizing objects and humans typically concentrates on a single technical solution in one of the following main categories: radio-based, acoustic, and image processing systems. Together with more mature systems (like GPS) numerous localization techniques coexist in isolation. But having a closer look on deploying localization systems in industrial applications several problems arise as the environmental influences vary too much to use a single technology. Moreover, the localization system needs to be integrated with other enterprise applications to enhance business processes. This work discusses how a localization system can help to improve occupational safety in industrial plants. We identified concepts to overcome the limitations of current stand-alone systems and evaluated the feasibility in a prototypical localization system, which we deployed in an industrial steel plant. This paper discusses how a localization system can help to improve occupational safety in industrial plants and presents concepts and findings gained during the deployment and test phase.

SemGen: towards a semantic data generator for benchmarking duplicate detectors

Benchmarking the quality of duplicate detection methods requires comprehensive knowledge on duplicate pairs in addition to sufficient size and variability of test data sets. While extending real-world data sets with artificially created data is promising, current approaches to such synthetic data generation, however, work solely on a quantitative level, which entails that duplicate semantics are only implicitly represented, leading to only insufficiently configurable variability. In this paper we propose SemGen, a semantics-driven approach to synthetic data generation. SemGen first diversifies real-world objects on a qualitative level, before in a second step quantitative values are generated. To demonstrate the applicability of SemGen, we propose how to define duplicate semantics for the domain of road traffic management. A discussion of lessons learned concludes the paper.