MLPaSSAD – New Multi-Layer Platforms for Security and Safety-Relevant Automated Driving Functions
Refine
Document Type
Language
- English (6)
Publication reviewed
- begutachtet (6)
Keywords
- MLPaSSAD (5)
- Advanced Driver Assistance Systems (2)
- Autoamted Driving (2)
- advanced driver assistance systems (2)
- automated driving (2)
- driving experience (2)
- human factor (2)
- human machine interface (2)
- lane keeping system (2)
- stress (2)
To ensure the safety and security of Automated Vehicles (Avs), the interaction between the Functional Safety (FuSa) and the Cybersecurity (CS) domains needs to be managed systematically. There is a demand to develop effective and structured management systems to support the homologation process. From this motivation, identifying the interaction between the Safety Management System (SMS) and the Cybersecurity Management System (CSMS) is a fundamental aspect and needs to be improved for HAD systems. Hence, the classical Decision Making Trial and Evaluation Laboratory (DEMATEL) method and fuzzy DEMATEL are applied to evaluate the influential factors that can impact the safety and security of the HAD systems. This paper proposes a list of influencing factors focusing on the interaction between SMS and CSMS for HAD systems. Additionally, the results of an anonymously conducted survey among experts from industry and research are presented and used as inputs for the methods. This work helps to understand the relationship between influencing factors and provides a simplified, easy-to-visualized, and valuable guide for developing HAD systems. The result of this study shows that the most important influential factor is F13. Moreover, the cause and effect of the factors are illustrated numerically and graphically. The influential factors F1 to F7 are identified as the cause and F8 to F13 are reasoned to effect. Finally, a circular representation of the influential factors and their interaction is presented in this paper.
Parking – Evaluation of Manual and Automated Parking Maneuvers with Subjective Assessment Indicators
(2021)
In this paper, an analysis of a subjective evaluation of manually and automatically executed longitudinal and lateral parking maneuvers using Subjective Assessment Indicators is presented. With the introduction of autonomous driving, parking maneuver assistants are essential functional components. Driver assistance systems will only be accepted if they perform decisively better than the human driver. Whether the performance of such a system meets expectations is ultimately a subjective impression. For this reason, an analysis of the parking performance of humans and parking assistant systems is carried out based on a new innovative subjective evaluation method. This new subjective evaluation method is based on the so-called Subjective Assessment Indicators which cover the relevant areas of a parking maneuver but still do not reach a level of detail that makes evaluation unsuitable for customer. Using the new subjective evaluation method, a driving study was conducted with 21 participants and two different test vehicles. The participants evaluated both manual and fully automated longitudinal and lateral parking maneuvers purely digitally using an evaluation app. As the results of the study show, parking assistants still have notable deficits compared to human performance in some evaluation areas and show considerable potential for improvement. As the subjective evaluation method used is suitable for all parking maneuvers and vehicle types, the results of this and potentially further studies form the basis for determining Key Performance Indicators for parking maneuvers. This enables virtual development of automated parking systems, as a link can be established to subjective customer evaluations.
The present study investigated the mental workload associated with driving a vehicle equipped with Lane Keeping Assistance System (LKAS). Specifically, an experiment was carried out with16participants driving with LKAS in four real-world scenarios. Effects on mental workload were evaluated with psychophysiological measures such as heart rate and skin conductance response (SCR). The driving performance, which is also a measure of evaluating mental workload, was assessed by measure such as steering reversal rate, variation of lateral position and steering effort. The result suggested that LKAS has reduced physical workload in the steering task. However, the lane keeping performance was not improved. Moreover, the NASA-TLX showed that participants perceived higher mental workload while driving with LKAS. This effect was mirrored in the SCR. The objective data showed that LKAS was associated with higher steering reversal rate, which might explain the reason of participants perceiving higher mental workload. Overall, it was suggested that the mental workload was higher with the tested LKAS.
In a recent study with N = 50 subjects, the lane keeping assistant was tested on more than 3,500 km on public roads in the Allgäu. To test the various settings of the lane keeping assistant, different conditions were tested: 120km/h versus 160km/h as well as with versus without lane keeping assistant. The evaluation of the criteria for lane keeping assistant, such as edge management and degree of relief show a significant relationship with the experienced workload. The increased workload as well as stress when using the lane keeping assistant system could be detected and proved subjectively as well as with physiological measuring devices. The significantly higher stress experienced with the use of the lane keeping assistant system shows the immense importance that the further research on this system has.
Deep learning for lateral vehicle control – an end-to-end trained multi-fusion steering model
(2019)
Deep Learning based behavior reflex methods found their way into modern vehicles. To model the human driving behavior it is not sufficient to rely solely on individual, noncontiguous camera frames without taking vehicle signals or road specific features into account. In this work four temporal fusion methods are evaluated based on three different Deep Learning models. The proposed spatio-temporal Mixed Fusion model extends the present end-to-end models and consist of multiple levels of fusions. The raw image data from a single front facing camera is mixed with recorded vehicle data and a map based predicted road bank angle gradient vector. The model accesses multiple time axes: temporal features of multiple image frames are extracted through a combination of Convolution and LSTM layers while it can also make assumptions about the future road condition with the use of upcoming Ground Truth road bank angle changes. Experiments are performed on a recorded data set of real world drivings. Results show, that this approach leads to an accurate imitation of the human driver with an inference capability of more than 60 FPS.