Marquart Franz

Variability in Foot-Worn Sensor Placement for Activity Recognition

On-shoe acceleration and orientation sensors have revealed as a potentially powerful means for capturing aspects of human gait. The placement of sensors however has been done intuitively and mostly without quantitative evaluation of sensor positioning. Based on recorded signals of the five placement options sole, heel, toe-cap, instep and ankle we built SVM classifiers using orientation-based features and evaluate their performance on three activity classes level walking, going upstairs and going downstairs. Finally we present an approach to a placement-invariant classification model and discuss the benefit for a bipedal sensing setup.

Building flexible manufacturing systems based on peer-its

Peer-to-peer computing principles have started to pervade into mechanical control systems, inducing a paradigm shift from centralized to autonomic control. We have developed a self-contained, miniaturized, universal and scalable peer-to-peer based hardware-software system, the peer-it platform, to serve as a stick-on computer solution to raise real-world artefacts like, for example, machines, tools, or appliances towards technology-rich, autonomous, self-induced, and context-aware peers, operating as spontaneously interacting ensembles. The peer-it platform integrates sensor, actuator, and wireless communication facilities on the hardware level, with an object-oriented, component-based coordination framework at the software level, thus providing a generic platform for sensing, computing, controlling, and communication on a large scale. The physical appearance of a peer-it supports pinning it to real-world artefacts, while at the same time integrating those artefacts into a mobile ad hoc network of peers. Peer-it networks thus represent ensembles of coordinated artefacts, exhibiting features of autonomy like self-management at the node level and self-organization at the network level. We demonstrate how the peer-it system implements the desired flexibility in automated manufacturing systems to react in the case of changes, whether intended or unexpectedly occuring. The peer-it system enables machine flexibility in that it adapts production facilities to produce new types of products, or change the order of operation executed on parts instantaneously. Secondly, it enables routing flexibility, that is, the ability to use multiple machines to spontaneously perform the same operation on one part alternatively (to implement autonomic fault tolerance) or to absorb large-scale changes in volume, capacity, or capability (to implement autonomic scalability).

Mobile Apps für industrielle Anwendungen am Beispiel von Siemens

„Smarte mobile Applikationen“, meist nur kurz Apps genannt, etablieren sich mehr und mehr zu einem eigenstandigen Zweig der Software-Entwicklung und tragen bereits heute substantiell zum Value Add einiger weniger Firmen wie beispielsweise Apple oder Google bei. Bislang sind die meisten Anwendungen jedoch dahingehend optimiert, einem einzelnen Anwender einen individuellen Mehrwert zu bieten, etwa anhand eines Reisefuhrers, durch den Zugriff auf ein soziales Netzwerk oder in der Form von Spielen fur den kurzfristigen Zeitvertreib. Fur einen weltweit tatigen Industriekonzern wie die Siemens AG ist es im Gegensatz dazu jedoch von weitaus groserem Interesse, mobile Anwendungen dabei als Teil des internen und externen Produktportfolios zu betrachten. Hier sind beispielsweise Applikationen zu nennen, welche mit den Siemens-eigenen Produkten eng zusammenarbeiten und diese sinnvoll erganzen konnen. In diesem Beitrag werden die speziellen Anforderungen an diese Klasse der industriellen Applikationen anhand der Automatisierungspyramide naher beleuchtet. Es werden anschliesend einige ausgewahlte Applikationen aus den Siemens-Sektoren vorgestellt, wobei die jeweils zugrunde liegende Problemstellung prasentiert und diskutiert wird. Ebenso werden erste Erfahrungen mit der jeweiligen App und der dadurch entstandene wirtschaftliche Nutzen erlautert. Abschliesend wird ein kurzer Ausblick auf die nachsten Schritte und die sich gerade in der Definition befindlichen Prozesse fur die App-Entwicklung innerhalb der Siemens AG gegeben.

Using Spatial Abstractions in Industrial Environments

The invisible integration of technology in literally everything, which is the fundamental of research on pervasive and ubiquitous computing, leads to huge quantities of smart objects which are distributed in physical space. They should operate autonomously, namely with as little human intervention as possible, and interact with humans in a more implicit way. Spatial properties and relations are important context information in this regard, as they allow them to automatically adapt to spatial changes in their environment in a semantically meaningful way. We propose novel concepts and a framework for exploiting qualitative abstractions of relationships between smart objects, which are specific to an application domain and facilitate the development of spatially-aware applications, and demonstrate their use with four industrial application scenarios.