IT-Anwenderzentrum (ITZ)
Refine
Document Type
Language
- English (14)
Has Fulltext
- no (14)
Is part of the Bibliography
- no (14)
Keywords
- Standards (3)
- Computer architecture (2)
- Kraftfahrzeug (2)
- Logic gates (2)
- Routing (2)
- Smart grids (2)
- Software (2)
- Softwarearchitektur (2)
- Survey (2)
- Algorithms (1)
Institute
Begutachtungsstatus
- peer-reviewed (5)
This paper focuses on Artificial Intelligence (AI) in robotic surgery. The question of safety and autonomy follows through the whole paper. Guidelines like Safety Integrity Levels which apply to dependable systems in general are described shortly. Overall, this work does not explicitly supply advantages of AI and instructional guidelines to build autonomous robots, instead, concentrates on challenges in the use of AI. In conclusion, there are still many open issues in the use of AI which cause potential gaps in reliability.
Considering the ongoing growth of Wireless Sensor Networks (WSNs) and the challenges they pose due to their hardware limitations as well as the intrinsic complexity of their interactions, specialized algorithms
have the potential to help solving these challenges. We present a survey on recent developments regarding algorithmic problems which have applications in wireless systems and WSNs in particular. Focusing on the intersection between WSNs and algorithms, we give an overview of recent results inside this intersection, concerning topics such as routing, interference minimization, latency reduction, localization among others. Progress on solving these problems could be potentially beneficial for the industry as a whole by increasing network throughput, reducing latency or making systems more energy-efficient. We summarize and structure these recent developments and list interesting open problems to be investigated in future works.
Performance Isolation of Co-located Workload in a Container-based Architecture for Automobile
(2019)
As the development in the automotive sector is facing upcoming challenges, the demand for in-vehicle computing power capacity increases and the need for flexible hardware and software structures arises, allowing dynamic managament of resources. In this new scenario, software components are to be added, removed, updated and migrated between computing units. To isolate the software components from each other and allow its orchestration, a container-based virtualization approach is being tested throughout this research. The analysis focuses on the question if this virtualization technology could be an option to ensure an interference-free operation. Four different sample applications from the automotive environment are tested for their susceptibility to resource contention. The research on the one hand shows that CPU and memory used by an application can
be largely isolated with this technology, but on the other hand, it
becomes apparent that support for I/O-heavy usage is currently
not implemented sufficiently for container engines.
Ontologies used in robotics
(2017)
Full autonomy of road vehicles is a major goal of the automotive industry. To reach such high autonomy it is necessary to provide an accurate and comprehensible situation description for the environment and the vehicle itself. A consistent depiction is essential to facilitate data exchange and communication between internal modules, e.g. collision check and environment model, as well as communication with further information sources like traffic participants vehicle to vehicle (V2V) or the infrastructure, e.g. smart traffic lights, road signs or radio traffic service (V2I). One necessary tool to create such a model could be an ontology which represents the given information and its dependencies. Initially this work provides a summary of given approaches in the literature for use of ontologies in robotics in general. Therefore approaches are stated sorted by application and task. Further approaches with focus on autonomous robots and in particular on autonomous vehicles are listed and described. Finally we give an outlook for further research topics in the domain of ontologies.
Assuming that future connected cars with automated driving functions will require even more computing power and communication bandwidth, the current network infrastructure as well as the existing individualized control units are not a profitable option for such vehicles. In addition, changing user expectations demand flexible architectural patterns and upgradeability of software components without the need to visit the workshop. However, the current statically developed and configured ECU architecture does not offer any practicable possibilities for this. For these reasons, the research for a new dynamic and flexible architecture is necessary. This new type of system architecture is expected to meet future requirements in terms of space, cost, performance, energy efficiency and number of required computing units in the vehicle, which will arise as a result of the implementation/inclusion of new automated driving functionalities, and due to the changes in user expectations.
Solutions to this issue can be found in the field of enterprise
IT (cluster computing), in which technologies such as Ethernet,
container-based virtualization and flexible software architectures have proven themselves to be very efficient for years.
Relevant infrastructures, for example from cloud computing providers, have commonly been used in high-performance or high-availability applications. Thus, in the research project A 3F has been investigated which of these concepts and methods can be applied to modern vehicle system architectures. One of the main goals is to assess the synergy potential of the two sectors, information technology and automotive industry, which to date have very different orientations. However, this synergy is expected to grow strongly in the course of the developments mentioned above.
In the following pages, the necessary changes related to hardware and software will be discussed briefly und will be compared to concepts and possible solutions from the IT world.
The vehicular ad hoc network (VANET) technology based on the approved IEEE 802.11p standard and the appendant inter-vehicle communication (IVC) has the potential to dramatically change the way transportation systems work. The fundamental idea is to change the individual behavior of each vehicle by exchanging information among traffic participants to realize a cooperative and more efficient ransportation system. Certainly, the evaluation of such systems is a comprehensive and challenging task in a real world test bed, therefore, simulation frameworks are a key tool to analyze IVC. Several models are needed to emulate the real behavior of a VANET in all aspects as much realistically as necessary. The intention of this survey is to provide a comprehensive overview of publications concerning IVC simulations of the year 2013 and to see how IVC simulation has changed since 2009. Based on this analysis, we will answer the following questions: What simulation techniques are applied to IVC? Which aspects of IVS have been evaluated? What has changed within five years of IVC simulations? We also take a closer look at commonly used software tools and discuss their functionality and drawbacks. Finally, we present open questions concerning IVC simulations.
Intelligent driver assistance systems and new infotainment innovations cause a rapidly growing demand of computing power. To satisfy that demand, the quantity of electronic control units in cars has increased dramatically. OEMs tackle that trend by consolidating software on powerful multicore hardware platforms. However, current software solutions are mostly static and designed to run on limited platforms. As promising operating system for automotive, Linux comes into consideration, which seems to scale better than already existing solutions. To ease the migration process of older software parts and guarantee freedom from interference according to ISO26262 between single software partitions, embedded hypervisors can achieve that requirements. Up to now, automotive systems are not developed to run within virtualized environments. Within this paper, we present an approach to map communication channels of virtual automotive ECUs and connect them with their already existing CAN interfaces. For our analysis, we use the Xen hypervisor. The focus for interaction between virtual machines is to use SocketCAN and given paravirtualized Ethernet drivers. Our goal is a non-intrusive software integration methodology. We keep the source code within software partitions as unmodified as possible. To benchmark our studies, we evaluate our implementation on the Intel i7 and the.
Smart Grid, Smart Metering, Electromobility and the regulation of the power network are keywords of the transition in energy policies. In the future, the power grid will be smart. Due to this the OTH Regensburg does research in this area based on different projects. This paper presents a data collection, analyzing and monitoring software for a reference Smart Grid. We discuss two possible architectures for collecting data from energy analyzers. The performance is analyzed with respect to real-time monitoring, analyzing load peaks and automated regulation of the power grid.
In this paper we present a practical example of the use of Cellular Communication standards like GPRS and UMTS in a Smart Grid Application. For a more detailed view we demonstrate a possible implementation of Cellular Communication Technologies in a data acquisition application for the collection of energy indicators in a medium-voltage grid. Furthermore, we show a technical overview of relevant and common mobile communication standards available in Germany. The included theoretical examples, Smart Grid scenarios, presented data and results are based on a research project for intelligent power regulation in medium-voltage grids performed in Regensburg (Germany). It is a joint project' of the University of Applied Sciences Regensburg together with a local energy provider and a manufacturer for distribution network systems.