Showing papers by "Ulrich Rückert published in 2003"

A massively parallel architecture for self-organizing feature maps

Abstract: A hardware accelerator for self-organizing feature maps is presented. We have developed a massively parallel architecture that, on the one hand, allows a resource-efficient implementation of small or medium-sized maps for embedded applications, requiring only small areas of silicon. On the other hand, large maps can be simulated with systems that consist of several integrated circuits that work in parallel. Apart from the learning and recall of self-organizing feature maps, the hardware accelerates data pre- and postprocessing. For the verification of our architectural concepts in a real-world environment, we have implemented an ASIC that is integrated into our heterogeneous multiprocessor system for neural applications. The performance of our system is analyzed for various simulation parameters. Additionally, the performance that can be achieved with future microelectronic technologies is estimated.

Method and arrangement for randomly storing data

Abstract: The invention relates to a method and an arrangement for randomly storing data in storage networks and/or an intranet and/or the internet, a corresponding computer program product, and a corresponding computer-readable storage medium, which are particularly suitable for distributing and retrieving data in error-tolerant and faulty systems such as storage networks or the internet. According to the inventive method for randomly storing data in storage networks and/or an intranet and/or the internet, one or several intervals, the total length of which corresponds to the relative capacity of the system, is/are assigned to each storage system. Said intervals are represented in a [0,1) interval but can overlap with other intervals as opposed to existing strategies. A real point is then assigned to each data block within the [0,1) interval by means of a (pseudo)random function. Optionally, said point can be part of several intervals of storage systems. A uniform placement strategy is used in order to assign the data block to one of said storage systems if that is the case. The interval lengths are adjusted correspondingly if the relative capacities of the storage systems change.

A holistic methodology for network processor design

Abstract: The GigaNetIC project aims to develop high-speed components for networking applications based on massively parallel architectures. A central part of this project is the design, evaluation, and realization of a parameterizable network processing unit. In this paper we present a design methodology for network processors which encompasses the research areas from the application software down to the gate level of the chip. Key components of this holistic approach have been successfully applied to characteristic examples of architecture refinements.

Directed power-variable infrared communication for the mini robot Khepera

Abstract: Communication facilities are important in Robotics if several robots have to work together. In this paper, we describe problems and solutions encountered while de- signing an infrared-based communication device for the mini robot Khepera. In contrast to traditional omnidirectional systems, it features directed, power-variable transmission in eight directions at 23.4 kbps up to a range of 1m . It can differentiate incoming data sig- nals from interference from adjacent sectors and can estimate their direction-of-arrival. We model the transmission over the infrared channel and show how interference influences the reception of the data signals. We also describe methods how to reduce these effects. We have tested the performance of the resulted signal processing in a worst case scenario by simulations and in experiments with a prototype implementation. The resulted module is especially suited for experimental evaluation of ad hoc network protocols and for position estimation.

A performance evaluation method for optimizing embedded applications

Abstract: Performance evaluation is an important step for designing embedded applications that require small footprints, low energy consumption and high throughput. We present a simulation-based method to characterize several resource properties (memory accesses, energy consumption, execution time) of embedded software that runs on dedicated processing engines targeted for SoC designs. The results of the characterization process are back-annotated to the source code to aid the designer in optimizing the implementation. Our approach allows the replacement of software parts by hardware units to speed up processing. We have performed case studies with software and hardware implementations of a pseudo-random number generator and a transmission error detector. The results show that computation speed-ups and energy reductions up to a factor of 15 can be obtained with implementations that exploit hardware extensions.

Gard – An Intelligent System for Distributed Exploration of Landmine Fields Simulated by a Team of Khepera Robots

Abstract: It is absolutely clear that state-of-the-art robots cannot undertake the whole procedure of terrorist bombs neutralization, unexploded ordnances clean up and minefield demining in many environmental situations, such as urban areas, but the main force toward building of a robotic system in these dangerous tasks is to reduce the human presence The study of robots for demining applications is indeed a scientifically challenging problem that offers wide possibilities of expanding the actual knowledge on several areas of robotics, ranging from localization devices to visual guidance systems, and from navigation on rough terrain to multi-agent co-operation The GPS and SLAM methods are the most efficient tools for a robot localization and mapping Their extension for a group of mobile robots is of an exceptional importance for the successful creation of the “landmines map” In this paper a co-operative localization algorithm using three mobile robots equipped with localization capabilities for detecting each other has been simulated and tested Further, navigation algorithms for a colony of mobile robots are proposed Results demonstrate the localization algorithm applicability, where with a low precision and restricted range, the odometry errors, which normally present problems for mapping, are severely reduced The simulations show that under certain conditions, successful localization is only possible if the team of multiple robots collaborate during localization by communication and data transferring

Storage Management as Means to cope with Exponential Information Growth

Abstract: The advances in Internet technology have led to tremendous improvements in business, education, and science and have changed the way we think, live, and communicate. Information exchange has become ubiquitous by the possibilities offered through modern technologies. We are able to offer information 24 hours a day through our web sites and can leave messages every time and from anywhere in the world. This change in communication has led to new challenges. Enterprises have to deal with an information amount that doubles every year. The technological foundation to cope with this information explosion is given by Storage Area Networks (SANs), which are able to connect a great number of storage systems over a fast interconnection network. However, to be able to use the benefits of a SAN, an easy-to-use and efficient management support has to be given to the storage administrator. In this paper, we will suggest new storage management concepts and we will introduce a new management environment that is able to significantly reduce management costs and increases the performance and resource utilization of the given SAN infrastructure.

Communication Between Khepera Mini Robots For Cooperative Positioning

Abstract: For implementing efficient cooperative behaviour strategies of a group of robots an appropriate communication system is needed. Depending on the robot’s task different communication techniques may be used. We have analyzed an ultrasonic detection and communication system, a communication technique that uses infrared transmitters and receivers and a radio based system using Bluetooth technology. In addition a passive vision system for cooperative robot recognition and positioning has been analyzed. For each system we have developed particular sensor and communication components, that can be used for experiments. As a platform for embedding and testing our hardware we use the mini robot Khepera.

Long-running Experiments using the Minirobot Khepera with Automatic Charging Station

Abstract: Ultrasonic sensors enable mobile autonomous systems to obtain information about obstacles in large environments. In the presented work, taking the limited energy resources of a mobile robot into account, only one transmitter is used. The employment of two receivers permits the calculation of the distance and direction of objects. Using pseudo-random sequences, simultaneously transmitting and receiving with the continuous perception of the environment is possible. The sequences are modulated in order to op- timise the signal-to-noise ratio and energy consumption with respect to the transducers’ bandwidth. With the gained signals the echo-intensity £eld is calculated for object detec- tion. Furthermore, several robots could operate in the same region applying orthogonal sequences.

Neural Object Classification Using Ultrasonic Spectrum Analysis

Abstract: This paper describes how reflected broadband sound signals are marked by interference phenomena if the surface of the ensonified object is structured. For an efficient extraction of features of the signal that relate to the surface structure of the object, the calculation of the cepstrum is introduced. A cylindrical test object is presented, which shows an angle independent and an angle dependent structure. This allows to specify the accuracy of the discrimination of structure sizes, that is based on a selected part of the cepstrum. In addition, the object can be used as a landmark for ultrasonic sensing. Classification of the cepstra is done by the statistical ‘One Nearest Neighbour’ (1NN) method as well as by a ‘Kohonen Self-Organising Feature Map’ (SOM). The results show, that changes in structure size down to 0.1mm are detectable.

Verfahren und anordnung zur randomisierten datenspeicherung

Abstract: Die vorliegende Erfindung beschreibt ein Verfahren und eine Anordnung zur randomisierten Datenspeicherung in Speichernetzwerken und/oder einem Intranet und/oder dem Internet sowie ein entsprechendes Computerprogramm-Erzeugnis und ein entsprechendes computerlesbares Speichermedium, welche insbesondere einsetzbar sind fur die Verteilung und das Wiederauffinden von Daten in fehlertoleranten sowie fehlerbehafteten Systemen. Hierfur wird vorgeschlagen, dass bei dem Verfahren zur randomisierten Datenspeicherung in Speichernetzwerken und/oder einem Intranet und/oder dem Internet jedem Speichersystem ein oder mehrere Intervalle zugeordnet werden. Diese Intervalle werden auf ein [0,1)-Intervall abgebildet, konnen sich aber im Gegensatz zu fruheren Strategien mit anderen Intervallen uberlappen. Jedem Datenblock wird nun mittels einer (pseudo- )zufalligen Funktion ein reeller Punkt im [0,1)-Intervall zugewiesen. Dieser Punkt kann zu mehreren Intervallen von Speichersystemen gehoren. Falls dem so ist, wird eine uniforme Platzierungsstrategie verwendet, um den Datenblock einem dieser Speichersysteme zuzuweisen. Verandern sich die relativen Kapazitaten der Speichersysteme, so werden die Intervalllangen angepasst.