TU Darmstadt / ULB / TUbiblio

Practical Feasibility and Functional Safety of a Wheeled Mobile Driving Simulator

Wagner, Paul (2018):
Practical Feasibility and Functional Safety of a Wheeled Mobile Driving Simulator.
Darmstadt, Technische Universität, [Online-Edition: https://tuprints.ulb.tu-darmstadt.de/7296],
[Ph.D. Thesis]

Abstract

The automotive development focus shifts from advanced driver assistance systems to-wards automated driving, as can easily be concluded from daily news. While the valida-tion and verification of driver assistance systems is already challenging, the question of how to validate automated driving is still unanswered. Nevertheless, it is widely agreed that the importance of Driving Simulators (DS) for the validation of driver assistance systems will increase even further for the validation of automated driving. Still, state-of-the-art DS are in a deadlock when it comes to providing the demanded quality in motion representation because larger workspaces are needed but cannot be provided economical-ly, thus, impeding validity of DS results and calling for a ground-breaking concept. A Wheeled Mobile DS (WMDS) is researched at FZD to replace state-of-the-art DS while providing an at least equal immersion to the test person with reduced costs. There-fore, this thesis investigates the superordinate project goal of proving feasibility of WMDS from two viewpoints: Firstly, is the wheeled motion base practically capable of providing the horizontal dynamics as they would occur in a real car in the aspects power demand, energy demand, and motion latency. Secondly, what measures are needed to reduce the risk that arises from the unbound system to an acceptable level and how are these measures triggered and monitored, ergo: How would a safety architecture need to look like for a WMDS? The first research question is addressed by conducting driving manoeuvres with the scaled WMDS prototype MORPHEUS. As unscaled urban driving manoeuvres cannot be driv-en with MORPHEUS, since the available driving areas are not large enough, a pow-er/energy model is developed, parameterised, and validated, enabling the simulation of the unscaled energy demand and of the power demand in dependence of the scaling factor. Concluding, the requirements to power and energy demand as well as motion latency can be fulfilled by state-of-the-art technology. To answer the second research question, a state-of-the-art hazard and risk analysis has been conducted and safety requirements have been derived. An overall safety architecture is designed for these safety requirements, whereas the core element is an autarchic emer-gency braking system. An exemplary design of this safety architecture is investigated and evaluated in terms of risk reduction and additional hazards arising from the newly intro-duced functions and components, yielding that no unacceptable risk is inherited in the system. Concluding, the herein presented work provides the next building block towards proving general feasibility of WMDS and, thus, towards revolutionising DS technology.

Item Type: Ph.D. Thesis
Erschienen: 2018
Creators: Wagner, Paul
Title: Practical Feasibility and Functional Safety of a Wheeled Mobile Driving Simulator
Language: English
Abstract:

The automotive development focus shifts from advanced driver assistance systems to-wards automated driving, as can easily be concluded from daily news. While the valida-tion and verification of driver assistance systems is already challenging, the question of how to validate automated driving is still unanswered. Nevertheless, it is widely agreed that the importance of Driving Simulators (DS) for the validation of driver assistance systems will increase even further for the validation of automated driving. Still, state-of-the-art DS are in a deadlock when it comes to providing the demanded quality in motion representation because larger workspaces are needed but cannot be provided economical-ly, thus, impeding validity of DS results and calling for a ground-breaking concept. A Wheeled Mobile DS (WMDS) is researched at FZD to replace state-of-the-art DS while providing an at least equal immersion to the test person with reduced costs. There-fore, this thesis investigates the superordinate project goal of proving feasibility of WMDS from two viewpoints: Firstly, is the wheeled motion base practically capable of providing the horizontal dynamics as they would occur in a real car in the aspects power demand, energy demand, and motion latency. Secondly, what measures are needed to reduce the risk that arises from the unbound system to an acceptable level and how are these measures triggered and monitored, ergo: How would a safety architecture need to look like for a WMDS? The first research question is addressed by conducting driving manoeuvres with the scaled WMDS prototype MORPHEUS. As unscaled urban driving manoeuvres cannot be driv-en with MORPHEUS, since the available driving areas are not large enough, a pow-er/energy model is developed, parameterised, and validated, enabling the simulation of the unscaled energy demand and of the power demand in dependence of the scaling factor. Concluding, the requirements to power and energy demand as well as motion latency can be fulfilled by state-of-the-art technology. To answer the second research question, a state-of-the-art hazard and risk analysis has been conducted and safety requirements have been derived. An overall safety architecture is designed for these safety requirements, whereas the core element is an autarchic emer-gency braking system. An exemplary design of this safety architecture is investigated and evaluated in terms of risk reduction and additional hazards arising from the newly intro-duced functions and components, yielding that no unacceptable risk is inherited in the system. Concluding, the herein presented work provides the next building block towards proving general feasibility of WMDS and, thus, towards revolutionising DS technology.

Place of Publication: Darmstadt
Divisions: 16 Department of Mechanical Engineering
16 Department of Mechanical Engineering > Institute of Automotive Engineering (FZD)
16 Department of Mechanical Engineering > Institute of Automotive Engineering (FZD) > Safety
16 Department of Mechanical Engineering > Institute of Automotive Engineering (FZD) > Test Methods
Date Deposited: 21 Oct 2018 19:55
Official URL: https://tuprints.ulb.tu-darmstadt.de/7296
URN: urn:nbn:de:tuda-tuprints-72966
Referees: Winner, Prof. Dr. Hermann and Prokop, Prof. Dr. Günther
Refereed / Verteidigung / mdl. Prüfung: 13 July 2018
Alternative Abstract:
Alternative abstract Language
Der automobile Entwicklungsfokus verschiebt sich von der Entwicklung von Fahrerassis-tenzsystemen (FAS) hin zu (hoch-)automatisiertem Fahren (HAF), wie der Tagespresse entnommen werden kann. Dabei sind sich Experten einig, dass Fahrsimulatoren (FS) bei der Absicherung von HAF eine noch wichtigere Rolle zukommen wird, also schon bei FAS. Der Stand der Technik der FS befindet sich in Bezug auf die Wiedergabequalität von Beschleunigungen jedoch in einer Sackgasse, da die gestellten Dynamikanforderun-gen nicht ökonomisch erfüllt werden können, was die Validität der im FS erzielten Ergeb-nisse beeinträchtigt und somit ein bahnbrechendes, neues Konzept auf den Plan ruft. Deshalb wird hier das Konzept eines selbstfahrenden Fahrsimulators (WMDS) unter-sucht, um bei mindestens gleichwertiger Immersion des Probanden und reduzierten Kos-ten den Stand der Technik zu ersetzen. Diese Arbeit beleuchtet das übergeordnete Pro-jektziel des Machbarkeitsnachweises von WMDS unter zwei Gesichtspunkten: Ist die selbstfahrende Plattform praktisch in der Lage die gleiche Horizontaldynamik, wie sie in einem realen Fahrzeug auftritt, in Bezug auf Leistungsbedarf, Energiebedarf und Bewe-gungslatenz abzubilden, sowie welche Funktionen und Überwachungsmaßnahmen sind erforderlich, um das von einem WMDS ausgehende Risiko auf ein akzeptables Niveau zu reduzieren? Die erste Forschungsfrage wird mittels Fahrversuchen mit dem skalierten WMDS-Prototyp MORPHEUS untersucht. Da aufgrund der begrenzten Fahrfläche für MORPHEUS keine unskalierten Stadtfahrmanöver durchführbar sind, wird ein Leis-tungs-/Energiemodell entwickelt, parametrisiert und validiert, mit dem ein unskalierter Energiebedarf sowie der Leistungsbedarf in Abhängigkeit des Skalierungsfaktors simu-liert werden. Der Leistungs- und Energiebedarf sowie die Anforderungen an die Bewe-gungslatenz sind nachweislich mit dem Stand der Technik erfüllbar. Die zweite Forschungsfrage wird untersucht, indem eine Gefahren- und Risikoanalyse durchgeführt wird und daraus Sicherheitsanforderungen abgeleitet werden. Eine Sicher-heitsarchitektur wird entworfen, wobei ein autarkes Notbremssystem das Kernelement darstellt. Eine exemplarische Ausführung der Architektur wird auf die erreichte Risikore-duktion sowie auf neuerlich generierte Gefahren durch die dem System hinzugefügten Komponenten und Funktionen untersucht und weist nach, dass keine unakzeptablen Risiken mehr vorhanden sind. Zusammenfassend liefert die vorliegende Dissertationsschrift den nächsten Baustein zum Machbarkeitsnachweis von WMDS und damit zur Revolution der FS-Technologie.German
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item