Otthein Herzog

WearIT@work: Toward Real-World Industrial Wearable Computing

Wearable computers are often cited as an enabling technology for out-of-office applications. In fact, there has been a considerable amount of work on industrial applications of wearables. However, with the notable exception of the symbol arm-worn system, this research has had little impact on industrial practice. The wearIT@work project is a 4 1/2-year effort financed by the European Union and aimed at facilitating real-life industrial deployment of wearable technology. The project is at the end of its third year. With 42 partners and a project funding of 23.7 million Euro (half of which comes from the EU), this consortium is the largest civilian wearable-computing effort worldwide. We are organizing the project around four pilot applications - aircraft maintenance, car production, healthcare, and emergency response - that drive the work in a bottom-up, user-centered approach.

Recognition and Prediction of Motion Situations Based on a Qualitative Motion Description

High-level online methods become more and more attractive with the increasing abilities of players and teams in the simulation league. As in real soccer, the recognition and prediction of strategies (e.g. opponent’s formation), tactics (e.g. wing play, offside traps), and situations (e.g. passing behavior) is important. In 2001, we proposed an approach where spatio-temporal relations between objects are described and interpreted in order to detect some of the above mentioned situations. In this paper we propose an extension of this approach that enables us to both interpret and predict complex situations. It is based on a qualitative description of motion scenes and additional background knowledge. The method is applicable to a variety of situations. Our experiment consists of numerous offside situations in simulation league games. We discuss the results in detail and conclude that this approach is valuable for future use because it is (a) possible to use the method in real-time, (b) we can predict situations giving us the option to refine agents actions in a game, and (c) it is domain independent in general.

Towards Ontology-Based Multiagent Simulations: The Plasma Approach.

In multiagent-based simulation systems the agent programming paradigm is adopted for simulation. This simulation approach offers the promise to facilitate the design and development of complex simulations, both regarding the distinct simulation actors and the simulation environment itself. We introduce the simulation middleware PlaSMA which extends the JADE agent framework with a simulation control that ensures synchronization and provides a world model based on a formal ontological description of the respective application domain. We illustrate the benefits of an ontology grounding for simulation design and discuss further gains to be expected from recent advances in ontology engineering, namely the adaption of foundational ontologies and modelling-patterns.

Towards Flexible and High-Level Modeling and Enacting of Processes

Process modeling and enacting concepts are at the center of workflow management. Support for heterogeneous processes, flexibility, reuse, and distribution are great challenges for the design of the next generation process modeling languages and their enactment mechanisms. Furthermore, flexible and collaborative processes depend also on unpredictable changes and hence require human intervention. Therefore, high-level process modeling constructs are needed which allow for an easy, adequate, and participatory design of workflows. We present a process modeling language which covers these requirements and is based on object-oriented modeling and enacting techniques. In particular, we outline how tasks and task nets are specified at a high level of abstraction, how flexible and user-adaptable control and data flow specifications are supported, and how reuse of workflow models can be improved. The approach is characterized by the uniform and integrated modeling of workflow schema and instance elements as objects and by the integration of flexible rule-based techniques with the high-level constructs of task graphs. Finally, we present our object-oriented approach for the distributed enactment of workflow models: A workflow is directly enacted by task agents which may be treated as reactive components, which interact by message passing, and whose execution behavior is derived from the context-free and context-dependent behavior of the tasks defined in the workflow schema.

Video retrieval by still-image analysis with ImageMiner

The large amount of available multimedia information (e.g. videos, audio, images) requires efficient and effective annotation and retrieval methods. As videos start playing a more important role in the frame of multimedia, we want to make these available for content-based retrieval. The ImageMiner-System, which was developed at the University of Bremen in the AI group, is designed for content-based retrieval of single images by a new combination of techniques and methods from computer vision and artificial intelligence. In our approach to make videos available for retrieval in a large database of videos and images there are two necessary steps: First, the detection and extraction of shots from a video, which is done by a histogram based method and second, the construction of the separate frames in a shot to one still single images. This is performed by a mosaicing-technique. The resulting mosaiced image gives a one image visualization of the shot and can be analyzed by the ImageMiner-System. ImageMiner has been tested on several domains, (e.g. landscape images, technical drawings), which cover a wide range of applications.

SMIZE: a spontaneous ride-sharing system for individual urban transit

We introduce a ride-sharing concept for short distance travel within metropolitan areas which is designed to handle spontaneous ridesharing requests of prospective passengers with transport opportunities available on short call. The system has been designed as a multiagent system. We present a methodology to determine the feasibility of our ride-sharing approach for specific metropolitan areas and predefined operation requirements using multiagent-based simulation. Concrete experiments have been conducted for the city of Bremen, (Germany) in the FIPA-compliant multiagent-based simulation system PlaSMA.

KI-98: Advances in Artificial Intelligence

Recently there has been a growing interest in the use of the biological immune system as a source of inspiration for solving complicated computational problems. The immune system involves many information-processing abilities including pattern recognition, learning, memory and inherent distributed p... Artificial Intelligence is the study of how to build or program computers to enable them to do what minds can do. This volume discusses the ways in which computational ideas and computer modeling can aid our understanding of human and animal minds. Major theoretical approaches are outlined, as well ...

Egocentric qualitative spatial knowledge representation for physical robots

Abstract Although recent (physical) robots have powerful sensors and actuators their abilities to show intelligent behavior is often limited. One key reason is the lack of an appropriate spatial representation. Spatial knowledge plays a crucial role in navigation, (self- and object-)localization, planning and reasoning for physically grounded robots. However, it is a major difficulty of most existing approaches that each of these tasks imposes heterogeneous requirements on the representation. In this paper, we propose an egocentric representation which relies on 1D ordering information that still provides sufficient allocentric information to solve navigation and (self- and object-)localization tasks. Furthermore, we claim that our approach supports an efficient, incremental process based on a simple 1D-representation.

Binarising SIFT-Descriptors to Reduce the Curse of Dimensionality in Histogram-Based Object Recognition

It is shown that distance computations between SIFT-descriptors using the Euclidean distance suffer from the curse of dimensionality. The search for exact matches is less affected than the generalisation of image patterns, e.g. by clustering methods. Experimental results indicate that for the case of generalisation, the Hamming distance on binarised SIFT-descriptors is a much better choice.

Automated interpretation and accessible presentation of technical diagrams for blind people

The EU-supported TeDUB (Technical Drawings Understanding for the Blind) project is developing a software system that aims to make technical diagrams accessible to blind and visually impaired people. It consists of two separate modules: one that analyses drawings either semi-automatically or automatically, and one that presents the results of this analysis to blind people and allows them to interact with it. The system is capable of analysing and presenting diagrams from a number of formally defined domains. A diagram enters the system as one of two types: first, diagrams contained in bitmap images, which do not explicitly contain the semantic structure of their content and thus have to be interpreted by the system, and second, diagrams obtained in a semantically enriched format that already yields this structure. The TeDUB system provides blind users with an interface to navigate and annotate these diagrams using a number of input and output devices. Extensive user evaluations have been carried out and an overall positive response from the participants has shown the effectiveness of the approach.