Now showing 1 - 10 of 83
  • Publication
    Towards Context-Awareness for Enhanced Safety of Autonomous Vehicles
    ( 2022)
    Haupt, Nikita Bhardwaj
    ;
    Autonomous vehicles operate in dynamic environments continuously encountering safety-critical scenarios. This necessitates employing methodologies that can handle these scenarios and ensure safety of the vehicle as well as other traffic participants. Besides, random failures or malfunctions in its components might result in hazardous situation(s), further raising concerns regarding safety. The intensity of these hazards caused by the malfunctions depends upon the current state of the operational context in which they occur. Thus to guarantee safe behavior of the vehicle, one must be aware of its operational context in the first place. To this end, we propose to systematically model the operational context of an autonomous vehicle apropos its safety-relevant aspects. This paper puts forth our initial work for context-awareness aided safety, including our perspective towards context and its modeling, and its categorization based on relevance and goal. We also propose a context meta-model and its fundamental elements crucial for developing a safety-relevant context model.
  • Publication
    Discovery of Perception Performance Limiting Triggering Conditions in Automated Driving
    ( 2021)
    Adee, Ahmad
    ;
    Gansch, Roman
    ;
    ;
    Glaeser, Claudius
    ;
    Drews, Florian
    Highly automated driving (HAD) vehicles are complex systems operating in an open context. Performance limitations originating from sensing and understanding the open context under triggering conditions may result in unsafe behavior, thus, need to be identified and modeled. This aspect of safety is also discussed in standardization activities such as ISO 21448, safety of the intended functionality (SOTIF). Although SOTIF provides a non-exhaustive list of scenario factors to identify and analyze performance limitations under triggering conditions, no concrete methodology is yet provided to identify novel triggering conditions. We propose a methodology to identify and model novel triggering conditions in a scene in order to assess SOTIF using Bayesian network (BN) and p-value hypothesis testing. The experts provide the initial BN structure while the conditional belief tables (CBTs) are learned using dataset. P-value hypothesis testing is used to identify the relevant subset of scenes. These scenes are then analyzed by experts who provide potential triggering conditions present in the scenes. The novel triggering conditions are modeled in the BN and retested. As a case study, we provide p-value hypothesis testing of BN of LIDAR using real world data.
  • Publication
    Wertschöpfung durch Software in Deutschland
    (Fraunhofer-Gesellschaft, 2021) ; ; ; ; ; ;
    Falk Howar
    ;
    ; ; ; ;
    Steffen, Barbara
    ;
    Nouak, Alexander
    ;
    Köhler, Henning
    Informationstechnologie und insbesondere Software ist ein wachsender Sektor in jeder entwickelten Gesellschaft. Softwarebasierte Produkte und Dienstleistungen sind die »digitale Infrastruktur« des 21. Jahrhunderts: Digitale Unternehmen, datenzentrierte Geschäftsmodelle und Dienstleistungen, maschinelles Lernen, Industrie 4.0, autonomes Fahren all diese Trends basieren auf leistungsfähigen Kommunikationsnetzwerken, modernen Rechenplattformen und Software- Stacks inklusive Basisdiensten, die derzeit von den sogenannten »Big Five« aus den USA dominiert werden: Google, Amazon, Microsoft, und in geringerem Maße, Apple und Facebook. Ein gesunder und wachsender IKT-Sektor ist die Basis für zukünftigen Wohlstand. Europa und auch Deutschland fallen hier in Bezug auf Innovation und Wachstum hinter die USA und Asien (China, Taiwan, Japan) zurück: Die 100 erfolgreichsten Softwareunternehmen stammen zu 90 Prozent aus den USA. Europa und der Rest der Welt importieren Leistungen dieser Unternehmen für den Betrieb der eigenen Infrastruktur. China dagegen baut mit Firmen wie Alibaba, Tencent und Baidu bereits ein eigenes unabhängiges Ökosystem auf, das digitale Infrastruktur (Online Handel, Cloud-Rechenplattformen, Soziales Internet, etc.) für die chinesische Gesellschaft und Wirtschaft bereitstellt. Insgesamt erzielen einige asiatische Länder Wachstumsraten ihrer IKT-Sektoren, die deutlich über dem Wachstum in Europa liegen. Bei der bereits erreichten und insbesondere bei der angestrebten Digitalisierung unserer Gesellschaft stellen diese beiden Sachverhalte strategische Risiken für unseren Wohlstand und unsere Unabhängigkeit dar. Bundeskanzlerin Angela Merkel kommentierte dies auf dem Digitalgipfel 2019 in Dortmund mit: »Europa muss das auch alles können! «Die skizzierte Lage ist in Zahlen und Analysen gut dokumentiert und allgemein akzeptiert. Die entscheidende Frage ist heute, wie und wo gehandelt werden kann und muss, um in Europa beziehungsweise in Deutschland die notwendige Stärkung und Unabhängigkeit des eigenen Software-Sektors zu erreichen. Dieser Bericht beleuchtet den Zustand des europäischen und deutschen Software-Ökosystems, analysiert potenzielle Risiken und Bedrohungen, insbesondere durch fehlende europäische Kompetenzen im Bereich Software- und Basisdienste. Zwar wird darauf Bezug genommen, wie die Ökosysteme in den USA und in China entstanden sind und wie sie gedeihen, eine erneute Gegenüberstellung von deutscher und US-amerikanischer oder chinesischer Softwareindustrie ist jedoch nicht Gegenstand des vorliegenden Papiers. Vielmehr ist es das Ziel, pragmatisch umsetzbare Handlungsempfehlungen für die Bundesregierung zur Erhöhung der softwarebasierten Wertschöpfung in Deutschland vorzustellen, die mit den bestehenden Stärken und mit der bestehenden Struktur der Wertschöpfung in Deutschland kongruent sind.
  • Publication
    Machine Learning Based Dynamic Risk Assessment for Autonomous Vehicles
    ( 2021)
    Patel, Anil Ranjitbhai
    ;
    Autonomous vehicles (AVs) are complex safety-critical systems that operate in an uncertain and dynamic environment. To ensure safety, Hazard Analysis and Risk Assessment (HARA) is recommended in ISO 26262. The entire process, however, is based on the very premise that a human driver is responsible for the safety of the vehicle. On the contrary, AVs function without any human intervention. Therefore, to ensure safe behavior in all possible situations, Dynamic Risk Assessment (DRA) at runtime is a necessity to make AVs aware of themselves about the current risk and take decisions accordingly; instead of relying on static worst-case assumptions. In this paper, we present a novel approach to identify and classify the severity and controllability rating class based on the measured data from the on-board sensors. Support Vector Machine (SVM) learning technique was used to train, test, and validate the model with diverse feature sets. We illustrate the presented approach by employing an example of adaptive cruise control and discuss the case study with initial findings.
  • Publication
    Continuous Systems and Software Engineering for Industry 4.0: A disruptive view
    ( 2021)
    Nakagawa, Elisa Yumi
    ;
    Antonino, Pablo Oliveira
    ;
    ; ;
    Context: Industry 4.0 has substantially changed the manufacturing processes, leading to smart factories with full digitalization, intelligence, and dynamic production. The need for rigorous and continuous development of highly networked software-intensive Industry 4.0 systems entails great challenges. Hence, Industry 4.0 requires new ways to develop, operate, and evolve these systems accordingly. Objective: We introduce the view of Continuous Systems and Software Engineering for Industry 4.0 (CSSE I4.0). Method: Based on our research and industrial projects, we propose this novel view and its core elements, including continuous twinning, which is also introduced first in this paper. We also discuss the existing industrial engagement and research that could leverage this view for practical application. Results: There are still several open issues, so we highlight the most urgent perspectives for future work. Conclusions: A disruptive view on how to engineer Industry 4.0 systems must be established to pave the way for the realization of the fourth industrial revolution.
  • Publication
    Industry 4.0 reference architectures: State of the art and future trends
    ( 2021)
    Nakagawa, Elisa Yumi
    ;
    Antonino, Pablo Oliveira
    ;
    ;
    Capilla, Rafael
    ;
    ;
    Industry 4.0 has led to a dramatic shift in manufacturing processes, which must be accomplished by interacting end-to-end industrial systems. While Industry 4.0 is still a big challenge for many manufacturing companies, reference architectures have been increasingly adopted in different domains to guide engineers on how their systems should interoperate and be structured. Companies have made different experiences with reference architectures for Industry 4.0. However, depending on the use cases addressed, a reference architecture may be more or less suited to support the transformation of a particular company. Besides, a complete understanding of existing representative architectures does not exist. The main goal of this work is to review existing reference architectures for Industry 4.0 and analyze them concerning their suitability for supporting Industry 4.0 processes and solutions. For this, we systematically researched these architectures and thoroughly analyzed and characterized them. We also address their use and technologies/tools that could support their implementation. As a result, we found that existing architectures still have a long way to go; hence, we present the most urgent steps for the near future. We conclude that the Industry 4.0 community is right in investing in reference architectures considering the future of Industry 4.0.
  • Publication
    Systematic Modeling Approach for Environmental Perception Limitations in Automated Driving
    ( 2021)
    Adee, Ahmad
    ;
    Gansch, Roman
    ;
    Highly automated driving (HAD) vehicles are complex systems operating in an open context. Complexity of these systems as well as limitations and insufficiencies in sensing and understanding the open context may result in unsafe and uncertain behavior. The safety critical nature of the HAD vehicles demands to model limitations, insufficiencies and triggering conditions to argue safe behavior. Standardization activities such as ISO/PAS 21448 provide guidelines on the safety of the intended functionality (SOTIF) and focus on the performance limitations and triggering conditions. Although, SOTIF provides a non-exhaustive list of scenario factors that may serve as a starting point to identify and analyze performance limitations and triggering conditions, yet no concrete methodology is provided to model these factors. We propose a novel methodology to model triggering conditions and performance limitations in a scene to assess SOTIF. We utilize Bayesian network (BN) in this regard. The experts provide the BN structure and conditional belief tables are learned using the maximum likelihood estimator. We provide performance limitation maps (PLMs) and conditional performance limitation maps (CPLMs), given a scene. As a case study, we provide PLMs and CPLMs of LIDAR in a defined scene using real world data.
  • Publication
    Uncertainty representation with extended evidential networks for modeling safety of the intended functionality (Sotif)
    ( 2020)
    Adee, Ahmad
    ;
    Munk, Peter
    ;
    Gansch, Roman
    ;
    Highly automated driving (HAD) vehicles are complex and safety critical systems. They are deployed in an intricate environment which undergoes continual changes. Complexity of these systems as well as sensing and understanding the operational environment results in uncertainties, which needs to be addressed for the safety of HAD vehicles. Ongoing standardization activities (ISO/PAS 21448) to provide Safety of the Intended Functionality (SOTIF) of HAD vehicles intend to address these issues. As part of the SOTIF argumentation, we propose a novel modeling method to represent uncertainty of the system and the environment as well as the propagation of uncertainty through the system. In our previous work, we classified three types of uncertainty, namely aleatory, epistemic and ontological for this purpose. In this paper, we provide multiple plausibility functions of Dempster-Shafer Theory to fully assimilate the representation of ontological uncertainty along with epistemic and aleatory. We implement our proposed method using a commercial Bayesian Network tool. We show the application of our method with a perception classification use case.
  • Publication
    Should we "safely" handle the uncertainties at runtime? - A rather seldom asked question
    Ipso facto ""Uncertainty is certain"" makes design and development of Cyber Physical Systems (CPS), specifically for safety critical scenarios, a challenging process. CPS are expected to function safely in unforeseen contexts, which are often characterized by the pervasive presence of uncertainty. There is a multitude of research and numerous approaches available for efficiently handling such uncertainties at runtime, but how many of them handle it from the viewpoint of safety assurance? Are the approaches which handle various possible uncertainties at runtime from safety assurance perspective need of the hour? This paper attempts to explore these issues and offers a rarely chosen but important perspective on handling uncertainties at runtime during the development of CPS. This paper is based on initial outcomes of an ongoing Systematic Literature Review (SLR) and consequent research on ""safe"" handling of uncertainties at runtime.