Skip to main content
main-content
Top

About this book

This book is the second volume reflecting the shift in the design paradigm in automobile industry. It presents contributions to the second and third workshop on Automotive Systems Engineering held in March 2013 and Sept. 2014, respectively.
It describes major innovations in the field of driver assistance systems and automated vehicles as well as fundamental changes in the architecture of the vehicles.

Table of Contents

Frontmatter

Development Process

Frontmatter

Chapter 1. Design of Ride Comfort Characteristics on Subsystem Level in the Product Development Process

In the automotive development process the significance of full vehicle ride comfort is becoming more important. Due to rising complexity and new boundary conditions upcoming in the development process, like a higher variety of models, higher functional demands, and decreasing development times, the design of respective ride comfort characteristics in early phases of the development is desirable. The necessity for a precisely defined and structured process is therefore increasing. In driving dynamics already a high progress is achieved in defining a respective process, which can be essentially attributed to the application of a subsystem level in the derivation of vehicle properties. In ride comfort however, the progress is less advanced, as no comparable subsystem methods or models exist. Therefore in the following the focus lies specifically on the integration of a subsystem level in the derivation process of vehicle properties from full vehicle to components. For that purpose, initially the automotive development process will be illustrated in its general structure and its specific realization in driving dynamics and ride comfort. The advantages and disadvantages of the respective disciplines will be emphasized. Furthermore the structure of subsystem models in ride comfort as well as associated concept parameters are introduced. In consideration of the new methodology, the integration within the automotive development process is illustrated and examples are given. Finally the findings of the investigation are summarized and the advantages of the methodology are emphasized.
Christian Angrick, Günther Prokop, Peter Knauer

Chapter 2. Methods for Change Management in Automotive Release Processes

The handling of changes in automotive release processes is a fundamental challenge of today’s development projects. This chapter examines strategies for the identification of the effects of changes and evaluates concepts for the estimation of resulting retest effort. It is determined that there exists no approach that is applicable for large systems at vehicle level and that allows a reliable selection of all tests necessary to analyze the impact of the change. To solve this problem, two general concepts for test selection techniques are proposed. Inclusion-based approaches identify tests from the set of not executed tests whereas exclusion-based approaches eliminate tests from the set of performed tests. The two concepts are compared via receiver operating characteristic and cost estimation. Furthermore, the exclusion-based test selection is described in detail. It offers the opportunity to reduce the automotive release effort without drawbacks in test quality.
Christina Singer

Requirement Analysis and Systems Architectures

Frontmatter

Chapter 3. Increasing Energy-Efficient Driving Using Uncertain Online Data of Local Traffic Management Centers

The main goals of today’s research and development are leading to different systems and topics for more energy-efficient technologies in powertrains and intelligent driver assistance systems. The funded project “Energieeffizientes Fahren 2014” (EFA 2014/2) aims for increasing the electric vehicles’ operation range. In order to reach this goal an approach has been chosen which includes infrastructure data using Vehicle-to-Infrastructure (V2I) communication technologies. Particularly traffic actuated traffic lights are being utilized since this is state of the art to optimize traffic flow. Based on the interaction between vehicle and infrastructure the driver will be able to achieve an energy-efficient manner of driving through additional information and integrated board aggregation. This approach has been successfully tested in Dresden.
Per Lewerenz, Günther Prokop

Chapter 4. Modelling Logical Architecture of Mechatronic Systems and Its Quality Control

In this work an integrated method for the development of mechatronic systems is presented for capturing information from requirements to code generation level, with derived and intermediated abstractions in a logical view. Our modelling theory, based on FOCUS, is a model-based engineering method for the development of reactive software systems. It supports the specific needs of the automation and automotive domains, and provides a model-based logical representation of the system together with a user-friendly integration of automatic verification. The scope of this work is to present our model-based development methodology for mechatronic systems, which provides an integrated way to define the respective engineering process models and formalisms, system requirements and architectures, to specify the behaviour of the system. Therefore, novel complementary analysis techniques can be applied, allowing the verification of properties, the validation of system design and derived model-based implementations. Moreover, a wider support for discipline neutral models reduces errors during integration of artefacts from individual disciplines.
Alarico Campetelli, Manfred Broy

Chapter 5. Functional System Architecture for an Autonomous on-Road Motor Vehicle

Autonomous driving is a widely discussed field of research with still growing interest. In addition to a lot of technical, legal and social questions to be solved, an immense challenge still remains in mastering the complexity of the resulting system which would eventually replace the driver. A supporting tool for developing complex systems is given by the functional system architecture, which describes the system on an abstract level independent of concrete technical solutions. Functional system architectures published in the context of autonomous driving do not cover all necessary functional requirements. However, they focus on different sub-aspects and functional mechanisms within this context.
Our functional system architecture, which has been developed in the research project Stadtpilot at the Technische Universität Braunschweig, focuses on systematization and a combination of localization- and perception-driven approaches into one single well-structured functional system architecture. It has been developed in a top-down approach based on a formulation of the functional requirements of an autonomous on-road motor vehicle, in the sense of a modular building block system. It covers the aspects of localization, environmental and self-perception, mission accomplishment, usage of map data and communication, and the integration of the human being as a passenger and as another traffic participant in the close surroundings of the autonomous vehicle.
Referring to our functional system architecture, we discuss some basic mechanisms of autonomous driving in the following article, which become transparent due to the architecture’s basic structure. Additionally, we discuss where current advanced driver assistance systems are located within this architecture. This makes the big efforts which still have to be made to fulfill the necessary functional requirements regarding an autonomous vehicle driving safely in public road traffic more transparent.
Richard Matthaei, Markus Maurer

Functional Safety and Validation

Frontmatter

Chapter 6. Towards a System-Wide Functional Safety Concept for Automated Road Vehicles

In this chapter, a process to derive a system-wide functional safety concept for automated road vehicles is presented and a short introduction of Skill and Ability Graphs for a functional safety concept is given. The process to develop a functional safety concept contains an extension to the ISO 26262 standard’s Driver Assistance System development process. This extension is a Skill Graph to model system skills in the concept phase. The Skill Graph improves the Hazard Analysis and Risk Assessment by modeling driving skills early in the development process. Additionally, the Skill Graph is transferred to an Ability Graph, used to design a self-perception and self-representation, which enables monitoring of the system’s operation and functional capabilities online. This self-representation can be part of a technical safety concept. Based on the ability levels, safety actions can be derived which maintain or reach a safe state of operation. As a result, a self-monitoring system is possible, in which humans, either aboard the vehicle or external, do not have to monitor the system.
Andreas Reschka, Gerrit Bagschik, Markus Maurer

Chapter 7. A Method for an Efficient, Systematic Test Case Generation for Advanced Driver Assistance Systems in Virtual Environments

In this chapter, a method for an efficient, systematic test case generation for the test of advanced driver assistance systems in virtual environments is presented. The method is one of four steps in a systematic test process. These four steps are (1) analysis of the system, (2) test case generation, (3) test execution, and (4) test evaluation. The analysis serves to identify factors that have an impact to the system. The aim of the test case generation is to discretize value-continuous parameters into equivalence classes and to reduce the number of test cases for necessary test coverage. The test case generation uses combinatorial algorithms to achieve this objective. A test case is generated based on a 4-level model, which consists of the road network, adjustments for special situations, dynamic elements, and environmental conditions. To generate reproducible test cases, a special control for dynamic elements is introduced to adapt the behavior of dynamic elements to non-deterministic target elements. The test case generation is presented in a case study of a constriction assist. The test evaluation is used to verify the system and to replay test cases or important factors to the previous steps of the test concept.
Fabian Schuldt, Andreas Reschka, Markus Maurer

Chapter 8. Validation and Introduction of Automated Driving

With the introduction of automated driving without driver supervision, the automotive industry breaks new ground not just in functionality, but also in terms of validation. Even the most extensive road tests cannot statistically prove the safety of an automated vehicle. This makes the use of alternative validation techniques necessary. In principle, these techniques are already known but in order to apply them many fundamentals must still be determined, especially the base amount of required field data. Thus, methods for obtaining this data from road testing and field applications gain a high importance for future safety certification as a basis for the approval of vehicle automation systems.
Hermann Winner, Walther Wachenfeld, Phillip Junietz
Additional information

Premium Partner

image credits