Functional Resonance Analysis in an Overtaking Situation in Road Traffic: Comparing the Performance Variability Mechanisms between Human and Automation

Automated driving promises great possibilities in traffic safety advancement, frequently assuming that human error is the main cause of accidents, and promising a significant decrease in road accidents through automation. However, this assumption is too simplistic and does not consider potential side effects and adaptations in the socio-technical system that traffic represents. Thus, a differentiated analysis, including the understanding of road system mechanisms regarding accident development and accident avoidance, is required to avoid adverse automation surprises, which is currently lacking. This paper, therefore, argues in favour of Resilience Engineering using the functional resonance analysis method (FRAM) to reveal these mechanisms in an overtaking scenario on a rural road to compare the contributions between the human driver and potential automation, in order to derive system design recommendations. Finally, this serves to demonstrate how FRAM can be used for a systemic function allocation for the driving task between humans and automation. Thus, an in-depth FRAM model was developed for both agents based on document knowledge elicitation and observations and interviews in a driving simulator, which was validated by a focus group with peers. Further, the performance variabilities were identified by structured interviews with human drivers as well as automation experts and observations in the driving simulator. Then, the aggregation and propagation of variability were analysed focusing on the interaction and complexity in the system by a semi-quantitative approach combined with a Space-Time/Agency framework. Finally, design recommendations for managing performance variability were proposed in order to enhance system safety. The outcomes show that the current automation strategy should focus on adaptive automation based on a human-automation collaboration, rather than full automation. In conclusion, the FRAM analysis supports decision-makers in enhancing safety enriched by the identification of non-linear and complex risks.

Multi-Cloud Resource Management Techniques for Cyber-Physical Systems

Information technology is based on data management between various sources. Software projects, as varied as simple applications or as complex as self-driving cars, are heavily reliant on the amounts, and types, of data ingested by one or more interconnected systems. Data is not only consumed but is transformed or mutated which requires copious amounts of computing resources. One of the most exciting areas of cyber-physical systems, autonomous vehicles, makes heavy use of deep learning and AI to mimic the highly complex actions of a human driver. Attempting to map human behavior (a large and abstract concept) requires large amounts of data, used by AIs to increase their knowledge and better attempt to solve complex problems. This paper outlines a full-fledged solution for managing resources in a multi-cloud environment. The purpose of this API is to accommodate ever-increasing resource requirements by leveraging the multi-cloud and using commercially available tools to scale resources and make systems more resilient while remaining as cloud agnostic as possible. To that effect, the work herein will consist of an architectural breakdown of the resource management API, a low-level description of the implementation and an experiment aimed at proving the feasibility, and applicability of the systems described.

Driver-like decision-making method for vehicle longitudinal autonomous driving based on deep reinforcement learning

Decision-making is one of the key parts of the research on vehicle longitudinal autonomous driving. Considering the behavior of human drivers when designing autonomous driving decision-making strategies is a current research hotspot. In longitudinal autonomous driving decision-making strategies, traditional rule-based decision-making strategies are difficult to apply to complex scenarios. Current decision-making methods that use reinforcement learning and deep reinforcement learning construct reward functions designed with safety, comfort, and economy. Compared with human drivers, the obtained decision strategies still have big gaps. Focusing on the above problems, this paper uses the driver’s behavior data to design the reward function of the deep reinforcement learning algorithm through BP neural network fitting, and uses the deep reinforcement learning DQN algorithm and the DDPG algorithm to establish two driver-like longitudinal autonomous driving decision-making models. The simulation experiment compares the decision-making effect of the two models with the driver curve. The results shows that the two algorithms can realize driver-like decision-making, and the consistency of the DDPG algorithm and human driver behavior is higher than that of the DQN algorithm, the effect of the DDPG algorithm is better than the DQN algorithm.

Human-like Decision-Making System for Overtaking Stationary Vehicles Based on Traffic Scene Interpretation

There are multifarious stationary vehicles in urban driving environments. Autonomous vehicles need to make appropriate overtaking maneuver decisions to navigate through the stationary vehicles. In literature, overtaking maneuver decision problems have been addressed in the perspective of either discretionary lane-change or parked vehicle classification. While the former approaches are prone to generating undesired overtaking maneuvers in urban traffic scenarios, the latter approaches induce deadlock situations behind a stationary vehicle which is not distinctly classified as a parked vehicle. To overcome the limitations, we analyzed the significant decision factors in the traffic scenes and designed a Deep Neural Network (DNN) model to make human-like overtaking maneuver decisions. The significant traffic-related and intention-related decision factors were harmoniously extracted in the traffic scene interpretation process and were utilized as the inputs of the model to generate overtaking maneuver decisions in the same manner with the human driver. The overall validation results convinced that the extracted decision factors contributed to increasing the learning performance of the model, and consequently, the proposed decision-making system enabled the autonomous vehicles to generate more human-like overtaking maneuver decisions in various urban traffic scenarios.

An end-to-end approach to autonomous vehicle control using deep learning

The objective of this work is to develop an autonomous vehicle controller inside Grand Theft Auto V game, used as a simulation environment. It is used an end-to-end approach, in which the model maps directly the inputs from the image of a car hood camera and a sequence of speed values to three driving commands: steering wheel angle, accelerator pedal pressure and brake pedal pressure. The developed model is composed of a convolutional neural network and a recurring neural network. The convolutional network processes the images and the recurrent network processes the speed data. The model learns from data generated by a human driver´s commands. Two interfaces are developed: one for collecting in-game training data and another to verify the performance of the model for the autonomous vehicle control. The results show that the model after training is capable to drive the vehicle as well as a human driver. This proves that a combination of a convolutional network with a recurrent network, using an end-to-end approach, is capable of obtaining a good driving performance even using only images and speed velocity as sensory data.

Improving the UX for Users of Automated Shuttle Buses in Public Transport: Investigating Aspects of Exterior Communication and Interior Design

As a precursor to future public transportation, automated shuttle buses can already be experienced in some test regions, but the general public still has reservations and may not yet be ready for this change. For example, the fact that such vehicles might operate independently (without a human driver) creates a barrier of uncertainty and mistrust among people. In this work, we aim to identify and classify the prevailing reservations and propose solutions. We followed the User Centered Design (UCD) process to design concepts that are specifically tailored to the needs of future public transport users. After related work analysis, on-site research, and pre-studies, two main studies were conducted specifically to address communication in the exterior (n = 24) and interior/service design (n = 21). For both studies, we applied a mixed-methods approach combining quantitative and qualitative measures. Our results indicate that, in general, existing ways of communication in the exterior are insufficient to meet future needs. The two visualization concepts for external communication developed in this work were rated (significantly) better in most dimensions of the User Experience Questionnaire (UEQ), when compared to the baseline condition with no additional visualization. Furthermore, preferences among the study participants towards simple, highly visible, and well-known lighting concepts could be observed. As for the interior, the results show that participants rated attractiveness highly for the two design concepts (closer, further in the future) as compared to current, state-of-the-art solutions (automated buses currently in operation). For the “near future” concept, the pragmatic quality dominated, while in the other (the “far future”) concept the hedonic quality was in the foreground. From the results, design recommendations in different categories were derived, which reflect the general openness of the public towards new technologies and interior approaches, but also point out the importance for privacy and designated personal spaces inside an (automated) shuttle bus. Some of the results do not strictly apply to automated shuttle buses, and can serve as valuable suggestions for improving conventional shuttle buses.

Development and Verification of Infrastructure-Assisted Automated Driving Functions

Automated vehicles we have on public roads today are capable of up to SAE Level-3 conditional autonomy according to the SAE J3016 Standard taxonomy, where the driver is the main responsible for the driving safety. All the decision-making processes of the system depend on computations performed on the ego vehicle and utilizing only on-board sensor information, mimicking the perception of a human driver. It can be conjectured that for higher levels of autonomy, on-board sensor information will not be sufficient alone. Infrastructure assistance will, therefore, be necessary to ensure the partial or full responsibility of the driving safety. With higher penetration rates of automated vehicles however, new problems will arise. It is expected that automated driving and particularly automated vehicle platoons will lead to more road damage in the form of rutting. Inspired by this, the EU project ESRIUM investigates infrastructure assisted routing recommendations utilizing C-ITS communications. In this respect, specially designed ADAS functions are being developed with capabilities to adapt their behavior according to specific routing recommendations. Automated vehicles equipped with such ADAS functions will be able to reduce road damage. The current paper presents the specific use cases, as well as the developed C-ITS assisted ADAS functions together with their verification results utilizing a simulation framework.

Modelling Automation–Human Driver Handovers Using Operator Event Sequence Diagrams

This research aims to show the effectiveness of Operator Event Sequence Diagrams (OESDs) in the normative modelling of vehicle automation to human drivers’ handovers and validate the models with observations from a study in a driving simulator. The handover of control from automation to human operators has proved problematic, and in the most extreme circumstances catastrophic. This is currently a topic of much concern in the design of automated vehicles. OESDs were used to inform the design of the interaction, which was then tested in a driving simulator. This test provided, for the first time, the opportunity to validate OESDs with data gathered from videoing the handover processes. The findings show that the normative predictions of driver activity determined during the handover from vehicle automation in a driving simulator performed well, and similar to other Human Factors methods. It is concluded that OESDs provided a useful method for the human-centred automation design and, as the predictive validity shows, can continue to be used with some confidence. The research in this paper has shown that OESDs can be used to anticipate normative behaviour of drivers engaged in handover activities with vehicle automation in a driving simulator. Therefore, OESDs offer a useful modelling tool for the Human Factors profession and could be applied to a wide range of applications and domains.

The Effects of Interruption on Takeover Performance in Automated Vehicles

As autonomous driving technology developing, the role of human driver becomes a passive passenger in an automated vehicle. Drivers would perform non-driving related tasks instead of driving, especially multitasking.. However, most of studies did not considered multitasking as NDRTs. In this study, experiment considering self-interruption is conducted and investigate effects of interruption on takeover performance. To investigate effects of interruption on takeover performance, experiment using driving simulator was conducted. Watching a short video was selected as a NDRT and there were three conditions of NDRT according to self-interruption: ‘Baseline’, ‘Monitoring condition’, and ‘Smartphone condition’. Takeover performance was measured by eyes-on time and deactivation time. There was no statistically significant difference for eyes-on time depending on interruption. However, interruption has a significant effect on deactivation time. Also, it was more effective to use a smartphone during interruption.