Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung IOSB

Filters

42
3
Reset filters

Settings
Now showing 1 - 10 of 42
  • Publication
    Towards Real-World Deployment of Reinforcement Learning for Traffic Signal Control
    ( 2021)
    Müller, Arthur
    ;
    Rangras, Vishal
    ;
    Ferfers, Tobias
    ;
    Hufen, Florian
    ;
    Schreckenberg, Lukas
    ;
    Jasperneite, Jürgen
    ;
    Schnittker, Georg
    ;
    Waldmann, Michael
    ;
    Friesen, Maxim
    ;
    Wiering, Marco
    Sub-optimal control policies in intersection traffic signal controllers (TSC) contribute to congestion and lead to negative effects on human health and the environment. Reinforcement learning (RL) for traffic signal control is a promising approach to design better control policies and has attracted considerable research interest in recent years. However, most work done in this area used simplified simulation environments of traffic scenarios to train RL-based TSC. To deploy RL in real-world traffic systems, the gap between simplified simulation environments and real-world applications has to be closed. Therefore, we propose LemgoRL, a benchmark tool to train RL agents as TSC in a realistic simulation environment of Lemgo, a medium-sized town in Germany. In addition to the realistic simulation model, LemgoRL encompasses a traffic signal logic unit that ensures compliance with all regulatory and safety requirements. LemgoRL offers the same interface as the well-known OpenAI gym toolkit to enable easy deployment in existing research work. To demonstrate the functionality and applicability of LemgoRL, we train a state-of-the-art Deep RL algorithm on a CPU cluster utilizing a framework for distributed and parallel RL and compare its performance with other methods. Our benchmark tool drives the development of RL algorithms towards real-world applications.
  • Publication
    Investigation in IoT and 5G architectures for deployment of Artificial Intelligence into urban mobility and production
    ( 2021)
    Ferfes, Tobias
    ;
    Müller, Arthur
    ;
    Schriegel, Sebastian
    ;
    Jasperneite, Jürgen
    The automation industry discusses the deployment of processing resources in applications since decades. The two major subjects of discussion are centralized and decentralized computing. The deployment of artificial intelligence (AI) applications and the processing power is facing the same issue nowadays. AI is used in many different areas like smart city, industrial automation for pattern recognition or as an expert system and so on. However, it remains a central question how to deploy the AI within an application and how to connect the AI to other devices, services and agents. In this paper we discuss state-of-the-art architecture to structure general IoT applications, as well as, the two most common processing concepts. Furthermore, use cases in urban mobility and industrial automation to deploy AI applications are presented and an In-depth view to the implementation is discussed. Pros, cons and problems of the implementations are shown, which can provide ideas and references when deploying an AI application.
  • Publication
    Interoperabilität von Plug-and-Work-fähigen OPC UA- und Cloud-basierten Smart Services
    ( 2019)
    Heymann, Sascha
    ;
    Weskamp, Jan Nicolas
    ;
    Pethig, Florian
    ;
    Stojanovic, Ljiljana
    ;
    Schriegel, Sebastian
    ;
    Watson, Kym
    ;
    Jasperneite, Jürgen
    Smart Services in a cloud enable secure data and service integration in cross-site application scenarios as well as 'plug and work' functions for devices, machines, and data analytics software by applying the industrial standards Open Platform Communications Unified Architecture (OPC UA) and Automation Markup Language (AutomationML or AML for short). OPC UA serves as a comprehensive and secure communication technology from the machine level into the cloud, thus enabling manufacturer-independent interoperability. A Cloud Coupler on the shop floor publishes the availability of a factory and selected process data to a service implemented on the cloud infrastructure. Customers or even smart machines can use this information about e.g. Resources, availability and price, to make decisions for placing an order and can track their orders in real time. In order to simplify the setup of the connection to the cloud and minimize the commissioning time the cloud coupler also aggregates all OPC UA servers on devices in the factory into a single OPC UA aggregated server address space. In this paper an architecture is proposed which uses a cloud coupler and Plug and Work techniques to register a new or retrofit factory in a cloud to use and share information in an interoperable way. It is shown that the use of standards reduces the effort needed for the definition of interfaces and integration of devices and software.
  • Publication
    Analysis of industrial production environments and derivation of a novel channel model towards optical wireless communication
    ( 2018)
    Schneider, Daniel
    ;
    Flatt, Holger
    ;
    Jasperneite, Jürgen
    ;
    Stübbe, Oliver
    Today radio based wireless communication technologies offer limited performance, whereas optical wireless com- munication systems (OWC) propose potentially a high performant, scalable communication system conforming to real time conditions. However, current studies imply, that OWCs still lack the necessary performance and robustness level for most wireless applications in industrial production environments. In this approach several types of noises for free-space optical communication systems are empirically analysed in an accredited, exemplary industrial production environment. While the channel noise is usually modelled by the signal to noise ratio it is found that real environments cannot be approximated by the usual static additive white gaussian noise. In this approach the accumulated measurement data represents the spectrum variation of different locations and times relating to different types of noise sources. The implementation in a total channel model allows the optimization of OWC designs like the channel access scheme or the modulation type concerning performance and robustness. Furthermore an additional measurement setup is proposed, capable of measuring and classifying existing noise sources in order to serve the design of OWC systems in industrial production environments.
  • Publication
    Assistance system to support troubleshooting of complex industrial systems
    ( 2018)
    Lang, Dorota
    ;
    Wunderlich, Paul
    ;
    Heinz, Mario
    ;
    Wisniewski, Lukasz
    ;
    Jasperneite, Jürgen
    ;
    Niggemann, Oliver
    ;
    Röcker, Carsten
    In ever changing world, the industrial systems become more and more complex. Machine feedback in the form of alarms and notifications, due to its growing volume, becomes overwhelming for the operator. In addition, expectations in relation to system availability are growing as well. Therefore, there exists strong need for new solutions guaranteeing fast troubleshooting of problems that arise during system operation. The approach proposed in this study uses advantages of the Asset Administration Shell, machine learning, and humanmachine interaction in order to create the assistance system which holistically addresses the issue of troubleshooting complex industrial systems.
  • Publication
    Reliability-oriented multipath communication for internet-based cyber-physical systems
    ( 2018)
    Elattar, Mohammad
    ;
    Friesen, Maxim
    ;
    Henneke, Dominik
    ;
    Jasperneite, Jürgen
    The Internet represents a cost-effective solution to realize the communications for Cyber-physical Systems (CPSs) that span large geographical areas. However, the reliability of today's Internet is inadequate to support, for example, smart grid applications with reliability requirements of 99-99.9999%. Therefore, the Reliable Multipath Communication for Internet-based CPSs (RC4CPS) approach was proposed. Unlike previous works, it supports such demanding applications. Nevertheless, it still lacks a protocol implementation to utilize it. Existing multipath protocols for the Internet such as Multipath TCP cannot be utilized directly for such applications. This is due to a number of reasons including the throughput-oriented nature of the protocols. In this paper, RC4CPS is integrated into the Parallel Redundancy Protocol for IP Networks (iPRP) to develop an easy-to-deploy reliability-oriented multipath protocol for Internet-based CPSs. The resulting protocol, called iPRP-RC4CPS, extends the original iPRP implementation to be used in the Internet and provides a dynamic multipath selection to cope with the varying nature of Internet paths. The obtained results show that iPRP-RC4CPS maintained 100% reliability during the evaluation interval.
  • Publication
    Utilizing Blockchain Technology in Industrial Manufacturing with the help of Network Simulation
    ( 2018)
    Kobzan, Thomas
    ;
    Biendarra, Alexander
    ;
    Schriegel, Sebastian
    ;
    Herbst, Thomas
    ;
    Müller, Thomas
    ;
    Jasperneite, Jürgen
    The fourth Industrial Revolution is finding its way into modern manufacturing sites. Assets with a certain degree of implemented communication technology will be enabled to interconnect and cooperate with practically every other entity via the Industrial Internet. An extreme amount of data and information will be exchanged all the time. Machines will be empowered to make crucial decisions autonomously influencing whole production processes. Erroneous, illegitimate or tampered data will lead to incorrect decisions and will be posing a huge threat to future strong cross-linked added value networks. An approach to tackle this issue comes from a technology called distributed ledger technology with its most known variant the ""blockchain"". Certain properties of blockchain technology are showing promising enhancements for industrial networks primarily in order to guarantee digital trust. In this paper, issues like the ability to scale and the adaption onto the requirements of industrial networks are investigated. It turns out that simulation approaches must be taken into consideration, and a first step for an implementation is presented.
  • Publication
    Investigation on a distributed SDN control plane architecture for heterogeneous time sensitive networks
    ( 2018)
    Schriegel, Sebastian
    ;
    Kobzan, Thomas
    ;
    Jasperneite, Jürgen
    Data-driven services (optimization or condition monitoring) are often deployed using cloud architectures. The shop floor itself becomes more and more flexible and reconfigurable using modular machine design and Plug and Play services. These industrial use cases induce additional requirements to communication systems: scalable real-time communication from sensor to cloud as well as seamless and automatic network configuration on the shop floor. A promising data plane technology for the Industrial Internet of Things (IIoT) is IEEE 802.1 Ethernet TSN that allows convergent and time sensitive communication. The configuration of the IIoT is complex because the IIoT is often large, growing and changing over time and often consists of heterogeneous network domains because of the brownfield and manifold requirements from the applications. Software-defined Networking (SDN) has the potential to reduce the engineering effort and to increase the operation efficiency (monitoring, diagnosis, reconfiguration) of heterogeneous IIoT. SDN Control Planes can be implemented as physical-central, logic-central, distributed or hybrid architecture. The different architectures have specific advantages and disadvantages regarding QoS, throughput and engineering efforts. A key role takes the East-West interface that handles the communication between distributed SDN Controllers. SDN Data Plane agents can help to manage legacy Data Planes with SDN. A TSN Nano Profile with inverse and cooperative operating Time Aware Shaper is a migration solution to upgrade legacy communication controllers with TSN functions. The disadvantages are less throughput and increased latency for acyclic traffic.
  • Publication
    Towards integration of industrial ethernet with 5G mobile networks
    ( 2018)
    Neumann, Arne
    ;
    Wisniewski, Lukasz
    ;
    Ganesan, Rakash SivaSiva
    ;
    Rost, Peter
    ;
    Jasperneite, Jürgen
    In contrast to the predecessors, the fifth generation of mobile networks (5G) will particularly address communication in production systems by providing machine-type communication and matching requirements regarding reliability and latency. However, 3GPP 5G will not be the one and only network technology that can be used to fulfill all the requirements raising from industrial applications at reasonable costs. It is more likely, that it complements other wired and wireless technologies, rather than replacing them. This paper presents different scenarios to combine 5G and Industrial Ethernet to a hybrid topology, it explains the characteristics and possible use cases for these scenarios and it sketches approaches to engineer and configure the hybrid network.