Jung, Rolf
Refine
Document Type
Publication reviewed
- begutachtet (23)
- nicht begutachtet (6)
Keywords
- SOTIF (5)
- ADS (3)
- Functional Safety (3)
- HADF (3)
- Highly Automated Driving (3)
- CarMaker (2)
- Cybersecurity Management System (2)
- DIN EN ISO 13849 (2)
- FuSa (2)
- HAD (2)
- Lane Change (2)
- Parameter variation (2)
- SMS (2)
- real time control (2)
- 3D Object Detection (1)
- ACC (1)
- ADAS (1)
- Automated Driving System (1)
- BPM (1)
- CNC controlled (1)
- CSMS (1)
- Collision (1)
- Collision detection (1)
- Cybersecurity management system (CSMS) (1)
- Decision Making Trial and Evaluation Laboratory (DEMATEL) (1)
- Deep Learning (1)
- Driver System Interactions (1)
- Driver Vehicle Interface (1)
- EC (1)
- FM (1)
- FS (1)
- Foreseeable Misuse (1)
- Functional Safety (FuSa) (1)
- Functional safety (FuSa) (1)
- Fuzzy DEMATEL (1)
- GSN (1)
- Gripper (1)
- HARA (1)
- HARA, HAZOP, SOTIF, FuSa, HAD, ADS (1)
- Highly Automated Driving (HAD) (1)
- Highly Automated Driving System (1)
- Highly automated driving systems (1)
- Intelligent gripper (1)
- LKAS (1)
- Lane change scenarios (1)
- LiDAR Point Cloud (1)
- MC (1)
- MPC (1)
- NHTSA (1)
- Optimization (1)
- PFC (1)
- Performance Evaluation (1)
- Proportional Integral Derivative (1)
- Radar (1)
- Reliability (1)
- Renewable Energy (1)
- Robotics (1)
- SN 29500 (1)
- SOTIF-related Use Case (1)
- STPA (1)
- Safety Management (1)
- Safety Management System (1)
- Safety management system (SMS) (1)
- Safety of the Intended Functionality (SOTIF) (1)
- Scenario reduction (1)
- Scenario-based testing (1)
- Simulation-Based (1)
- Simulation-based testing (1)
- Statistical Analysis (1)
- Swarm Intelligence (1)
- TARA, OTA, HADF, FuSa, SOTIF. (1)
- TGAS (1)
- Trajectory Generation (1)
- embedded (1)
- gripper (1)
- life expectancy (1)
- machine control (1)
- optical (1)
- precision (1)
- reactive ion etcher (1)
- retrofit (1)
- safety (1)
- software architecture (1)
Institute
Scenario-based collision detection using machine learning for highly automated driving systems
(2023)
Highly Automated Driving (HAD) systems implement new features to improve the performance, safety and comfort of partially or fully automated vehicles. The identification of safety parameters by means of complex systems and the driving environment is a fundamental aspect that require great attention. Therefore, much research has been conducted in the field of collision detection in the development of automated vehicles. However, the development of HAD systems faces the challenge of ensuring zero accidents. For this reason, collision detection in the safety-related concept phase as hazard identification is one of the key research points in HAD system. In this paper, a systematic approach to detect potential collisions for scenario-based hazard analysis of HAD systems is presented by using Multilayer Perceptron (MLP) as a Machine Learning (ML) technique. Moreover, the proposed approach assists in reducing the number of observed scenarios for hazard analysis and risk assessment. Additionally, two simulation-based scenario datasets are examined in the ML model to identify potential hazard scenarios. The results of this study show that MLP can support to detect the collision at safety-related concept phase. Furthermore, this paper contributes to providing arguments and evidence for ML techniques in HAD systems safety by selecting relevant use cases.
To ensure safety and security of highly automated driving systems one shall make sure all risks are reduced to a reasonable level and an all potential cyberattacks are addressed with necessary protection. Because of the complexity of such vehicle systems, systematic and structured management approaches are vital to maintaining safety via cybersecurity (CS). The interface of Safety Management System (SMS) with Cybersecurity Management System (CSMS) is one of the key aspects to ensuring that potential safety issues are addressed. Both management systems include planning, concepts, and process development, with significant areas of overlapping management systems is required. Regarding the management systems interface and distribution, it is still a challenge that Highly Automated Driving (HAD) vehicles needs to overcome by means of effective implementation and strategies with continuous improvement and a reduction of miscommunication. From that motivation, a set of engineering risk management framework are proposed in this paper. Subsequently, introducing the interface areas between the safety and the cybersecurity domain is one of the focus areas of this paper, together with the representation of the interface management activities with exemplary interaction template. Additionally, mapping in between safety and cybersecurity related standards in terms of evidence and management systems is represented partially to support both safety case and security assurance.
Simulation-Based Testing of Foreseeable Misuse by the Driver Applicable for Highly Automated Driving
(2022)
With Highly Automated Driving (HAD), the driver can engage in non-driving-related tasks. In the event of a system failure, the driver is expected to reasonably regain control of the Automated Vehicle (AV). Incorrect system understanding may provoke misuse by the driver and can lead to vehicle-level hazards. ISO 21448, referred to as the standard for Safety of the Intended Functionality (SOTIF), defines misuse as the usage of the system by the driver in a way not intended by the system’s manufacturer. Foreseeable Misuse (FM) implies anticipated system misuse based on the best knowledge about the system’s design and the driver’s behavior. This is the underlying motivation to propose simulation-based testing of FM. The vital challenge is to perform simulation-based testing for a SOTIF-related misuse scenario. Transverse Guidance Assist System (TGAS) is modeled for HAD. In the context of this publication, TGAS is referred to as the “system,” and the driver is the human operator of the system. This publication focuses on implementing the Driver-Vehicle Interface (DVI) that permits the interactions between the driver and the system. The implementation and testing of a derived misuse scenario using the driving simulator ensure reasonable usage of the system by supporting the driver with unambiguous information on system functions and states so that the driver can conveniently perceive, comprehend, and act upon the information.
Scenario analysis is essential for the validation of highly automated driving (HAD) systems. The complexity of overall system safety is increasing in terms of Functional Safety (FuSa) and Safety of Intended Functionality (SOTIF). However, field testing of all possible safety-critical scenarios is hardly possible for automated vehicles. Therefore, scenario simulation is necessary for HAD to support the validation process and has gained acceptance in recent years. However, scenariobased analysis leads to an explosion of scenarios, so a scenario database is required at the beginning of the development phase. Hence, scenario reduction approaches need to be integrated in the conceptual phase to reduce the scenario modeling and testing effort. The contribution of this paper is to present simulation-based testing approaches for determining a reduced set of collision scenarios, taking into account the sensitivity of safety-critical parameters. Furthermore, a set of application scenarios is simulated to demonstrate the scenario reduction approaches by considering the detected collisions under a specific or restricted Operational Design Domain (ODD). In addition, this study supports the provision of a safety argument with evidence by introducing a variance-based sensitivity analysis and a scenario database that can be used as an input data set for an artificial intelligence system.
Modeling and simulation techniques are a necessity to solve the problems and aid the automated driving verification and validation process. The scenario-based analysis like hazard analysis and risk assessment is counting as an essential method not only to understand the system behavior in the field of an automated vehicle but also to reduce the development and communication gaps. In terms of functional safety and safety of the intended functionality the number of hazardous scenarios increases that need to be reduced. Scenario reduction is a challenge that yet needs to be solved. Therefore, this paper proposes a probability approach like the Monte Carlo method at the logical scenario level. Additionally, the safety-critical vehicle parameter range has been optimized based on collision detection. Furthermore, the result realized by the Monte Carlo experiment has been used to model the concrete scenarios in CarMaker in a time-efficient manner. The approach of modeling for a specific function like transverse guidance can be utilized to build a full scenario database for the highly automated driving vehicle.
To allow a vehicle with highly automated driving functions to operate on the road the overall safety (safe functionalities, functional insufficiencies including cybersecurity) of the driving system must be guaranteed. Therefore, a generic Safety Management System (SMS), that includes all useful and necessary regulations should be applied. The given specifications regarding the safety of driving systems shall be understood and considered in order to define an acceptable SMS for the Highly Automated Driving Function (HADF). Derived from the generic SMS a specific management system has to be developed to guide the development and deployment of the HADF. The research presented in this paper investigates the currently available SMS in different sectors like aviation, marine, and railway to propose a new set of components and elements that are useful and modified for a HADF’s SMS. Moreover, the paper provides a systematic approach to how the new set of components and elements can be applied for a HADF SMS. Additionally, well established hazard identification methods (scenario-based HARA and STPA) are compared and integrated into the safety concept phase. The complexity of scenarios is unique for a HADF. Since the SMS is going to be complicated for automotive HADFs and a constantly developing process, methods for evaluation and continuous improvement are needed. This paper gives insight into the structure of the SMS and to guarantee its applicability the use of a helpful software tool is considered. Furthermore, the proposed SMS approach can be adjusted as a groundwork for research concerns like validation for homologation and assess the safety of a HADF.
One of the fundamental tasks of autonomous driving is safe trajectory planning, the task of deciding where the vehicle needs to drive, while avoiding obstacles, obeying safety rules, and respecting the fundamental limits of road. Real-world application of such a method involves consideration of surrounding environment conditions and movements such as Lane Change, collision avoidance, and lane merge. The focus of the paper is to develop and implement safe collision free highway Lane Change trajectory using high order polynomial for Highly Automated Driving Function (HADF). Planning is often considered as a higher-level process than control. Behavior Planning Module (BPM) is designed that plans the high-level driving actions like Lane Change maneuver to safely achieve the functionality of transverse guidance ensuring safety of the vehicle using motion planning in a scenario including environmental situation. Based on the recommendation received from the (BPM), the function will generate a desire corresponding trajectory. The proposed planning system is situation specific with polynomial based algorithm for same direction two lane highway scenario. To support the trajectory system polynomial curve can be used to reduces overall complexity and thereby allows rapid computation. The proposed Lane Change scenario is modeled, and results has been analyzed (verified and validate) through the MATLAB simulation environment. The method proposed in this paper has achieved a significant improvement in safety and stability of Lane Changing maneuver.
For a safe operation of highly automated driving (HAD) systems on the road, the overall safety (functional safety (FS) and the safety of the intended functionality (SOTIF)) including cybersecurity must be guaranteed. Within these terms, the application of a Safety Management System (SMS) is becoming essential for HAD systems. Meanwhile, the cybersecurity for HAD systems is handled separately, familiar as Cybersecurity Management System (CSMS). To provide seamless safety and cybersecurity for highly automated vehicles on the road the two Management Systems should be linked with each other and therefore the interfaces from the SMS to cybersecurity must be defined. Furthermore, the communication flow from SMS to CSMS shall be specified to ensure that no information is lost in between. However, the development phases (SMS and CSMS) will also seemingly pomp the interface areas to industrial sectors like IEC TR 63069 [1], with UN Regulation No.155 [2] and others. The research aspects presented in this paper focus on:
(1) identifying the management related adaption topics for HAD-Safety and cybersecurity considering ISO 26262 (FS), ISO/PAS 21448 (SOTIF), UL4600 (safety for evaluation), SAE J3061 (cybersecurity guidebook) and upcoming ISO 21434 (cybersecurity) [3], [4] [5], [6] and [7].
(2) the novelty in interfaces of SMS and CSMS in risk management of HAD systems.
Moreover, provides a brief systematic approach to sort the collected data during the SMS processes into safety, cybersecurity, or both domains. Safety considers the functional insufficiencies (SOTIF) for hazard analysis and cybersecurity contemplates the threat analysis which can have the potential to occur a hazardous situation for HAD vehicles. Safety considers the SOTIF aspect which deals with environmental and misuse aspects but cybersecurity covers a vast area of consideration. Derived from the proposed approach a general structure for SMS interfaces to cybersecurity is created.
Advanced Driver Assistance System (ADAS) is playing a vital role in the development of human life. Understanding and identifying the necessity of Functional Safety (FuSa) for automated driving systems (ADS) is the focus of this presentation.
Several assist systems like ACC, LDW is now in demand from vehicle user because of comfort and safety. Since current standards focus on systems that are controlled by humans it is very important to clarify which of the given regulations are appropriate to define a common safety management system for automated vehicles. Besides, it is necessary to point out the gaps and missing topics that have to be recorded in specifications to provide a safe operation of tomorrow's vehicles.
Highly automated driving systems perceive the necessity of implementing safety and security features. In the presentation, Automation level 3 or higher is considered for implementing safety including security. Additionally, the necessity of the FuSa for ADAS/ADS is established and gives a glimpse of the research aspect in this area.
The development of a safety case for an automated driving system (ADS) with highly automated driving (HAD) functions is a flourishing area of research. One of the research sections is considering ethics commission (EC) rules with Functional Safety (FuSa), Safety Of The Intended Functionality (SOTIF) and Cybersecurity for ADS. A new Method should be able to provide the evidence to build confidence and ensure the robustness of ADS. Pegasus has already presented an approach for the representation of the logical structure of safety argumentation but has fallen short of applying it by actually modelling the ethic rules (or other safety princples like NHTSA, UL4600, etc) into the structure and connecting them with a concrete ADS safety case. Moreover, the representation of atomic consideration of design principles and their interrelation with ethical rules is incapacitated. Therefore, an elaborate approach will be presented (based on the Pegasus method), to show the relation of Safety (FuSa, SOTIF, cybersecurity and others) with ethical aspects in the development of ADS vehicle and support to build a strong safety argumentation for an ADS safety case.
Firstly, a set of atomic claims from both ethical rules and ADS design and safety principles is generated. Secondly, based on goal structuring notation (GSN), a subsequent safety argumentation and evidence is demonstrated from the generated claims for a dedicated HAD transverse guidance funtion from the ethical aspect. Finally, it is shown how consulted claims reflect safety and how safety can fulfil the ethical goals for ADS.