Accurately and proactively alerting drivers or automated systems to emerging collisions is crucial for road safety, particularly in highly interactive and complex urban environments. Existing methods require labour-intensive annotation of sparse risk, struggle to consider varying contextual factors or are tailored to limited scenarios. Here we present the generalized surrogate safety measure (GSSM), a data-driven approach that learns collision risk from naturalistic driving without the need for crash or risk labels. Trained on diverse datasets and evaluated on 2,591 real-world crashes and near-crashes, a basic GSSM using only instantaneous motion kinematics achieves an area under the precision–recall curve of 0.9 and secures a median time advance of 2.6 s to prevent potential collisions. Incorporating more interaction patterns and contextual factors provides further performance gains. Across interaction scenarios, such as rear end, merging and turning, GSSM consistently outperforms existing baselines in terms of accuracy and timeliness. These results establish GSSM as a scalable, context-aware and generalizable foundation for identifying risky interactions before they become unavoidable and support proactive safety in autonomous driving systems and traffic incident management.

Drivers often misjudge the capabilities of Advanced Driver Assistance Systems (ADAS), compromising safety. Guided by a Context–Vehicle–Driver (C-V-D) framework drawn from 22 empirical studies, this study analyzed a secondary survey of 838 drivers to identify predictors of self-reported “ADAS unawareness” (“I don’t know if I use it”). Analysis of the representation ratio (RR) showed that drivers with a low annual driving distance (' 5000 km), lack of private car ownership, and young age (18–29 years) were consistently overrepresented among unaware users (RR ≥ 1.2), while car sharing frequency and license tenure were not. Un-awareness was highest for Adaptive Cruise Control (ACC) among the three ADAS examined. These results support a hierarchical account in which contextual factors outweigh vehicle and driver-level influences. The C-V-D model yields testable hypotheses for road type, traffic density, and interface design that merit evaluation in larger-sample studies. Addressing the priority groups identified here can help designers, dealers, and educators reduce mode confusion and promote safe ADAS adoption.

Lane change modelling is a key challenge in driver behaviour research, as lane-changing involves complex interactions with surrounding vehicles and significantly impacts traffic flow. Despite advancements in lane change modelling, calibration is often limited to macroscopic indicators, limiting microscopic performance assessment. Yet, micro-level accuracy is crucial, in particular for simulating mixed traffic and setting up realistic testing environments for automated driving functions. Hence, this study evaluates the micro-level performance of existing lane change models (LMRS and MOBIL) by utilising naturalistic trajectory data. Additionally, comprehensive data processing aspects are addressed, and trajectory-based model parameter optimisation is conducted, enabling manoeuvre-level performance evaluation. The results reveal that MOBIL struggles to capture anticipatory human driver behaviour. This issue arises because MOBIL perceives close but faster vehicles in the target lane as obstacles, whereas in reality, drivers often do not treat them as such. LMRS, on the other hand, shows difficulties with gap acceptance.

Automated vehicles (AVs) promise to enhance transportation safety and efficiency. However, ensuring their reliability in real-world conditions remains challenging, particularly due to rare and unexpected situations known as edge cases. While numerous approaches exist for detecting edge cases, a comprehensive survey reviewing these techniques is lacking. This paper bridges this gap by presenting a hierarchical review and systematic classification of edge case detection and assessment methodologies. Our classification is structured on two levels: first, by AV modules, including perception and trajectory-related (encompassing prediction, planning, and control) subsystems; and second, by underlying methodologies and theories guiding these techniques. Furthermore, we introduce “knowledge-driven” approaches, which complement data-driven methods by leveraging expert insights and domain knowledge to identify cases absent in training datasets. We then examine techniques and metrics for evaluating edge case detection methods, including detection performance, practical deployment (e.g., computational overhead), and domain-specific measures (e.g., crash rates and severity analysis). We conclude by highlighting key challenges for edge case detection, including data availability and quality issues, validation and interpretability limitations, the simulation-to-real gap, and computational constraints. The hierarchical classification and review of methods and assessment techniques in this survey enable modular and targeted testing frameworks by guiding the selection of detection methods for specific AV subsystems while considering methodological principles. It also supports practical testing by facilitating scenario generation in simulation and focused subsystem validation in the real world.

Many drivers misjudge what their vehicle’s automation systems can actually do. This mismatch, known as mode confusion, can turn small misunderstandings into fatal consequences. Research has long examined drivers’ mental models and drivers’ confidence in engaging Advanced Driver Assistance Systems (ADAS), treating both as key contributors to mode confusion. Yet one crucial question remains largely unaddressed: do drivers know, correctly and confidently, which automation features are installed in their own vehicles? To address this question, we surveyed 1,487 U.S. vehicle owners whose manufacturers list Adaptive Cruise Control (ACC) and Lane Keeping Assistance (LKA) as standard equipment. Each respondent’s self-reported ownership awareness was compared with external model-trim data. Despite generally high ownership confidence, 17.1% incorrectly believed their vehicle lacked ACC and 29.4% believed it lacked LKA. Ownership awareness is uneven across ADAS: LKA is misjudged more often than ACC, even among drivers who are confident in their ownership judgments. Specifically, owning an older vehicle is associated with lower ownership accuracy and lower ownership confidence, while exposure to demanding trip contexts is more strongly related to lower ownership confidence than to lower ownership accuracy. Analyses of self-reported reasons using Holm-adjusted Fisher tests and association-rule mining reveal why ownership-awareness misalignment occurs. Misaligned ownership awareness commonly co-occurs with a lack of information and a lack of first-use experience, often coupled with an acceptance barrier that may reflect reluctance to engage initially with ADAS. In contrast, correct-and-confident ownership awareness co-occurs with prior ADAS use, clear in-vehicle feedback, and dealer explanation. Taken together, our findings suggest opportunities to help mitigate early mode confusion, including enhancing feedback and status visibility in in-vehicle interfaces and supporting guided first use through sales interactions or in-vehicle onboarding experiences, both of which warrant further testing.

Noise pollution negatively affects health and well-being, making its monitoring important for effective mitigation strategies. Sensor systems such as sound level meters have long been used for this purpose. Nevertheless, dependence on grid power, restricted metrics beyond loudness, and high costs per unit limit current solutions. This paper presents an open hardware, off-grid sound sensor to measure loudness and complementary noise metrics. The sensor detects eleven common urban sound events, calculates acoustic sharpness, and the intermittency ratio of the acoustic environment. The sensor is based on an ESP32-S3 microcontroller on a customized printed circuit board, optimized to address the current limitations. The board includes a battery management circuit for solar charging, a real-time clock for accurate time keeping, and supports LoRaWAN to send aggregated metrics. The latter allows remote monitoring, while more detailed metrics are stored on a microSD card. A solar panel and up to two 18650 Li-Ion or LiFePo4 batteries allow the sensor to be deployed independently of mains power. The open hardware is accompanied by open firmware, which has been organized into multiple components to allow easy changes and extensions for other use cases. A lab validation showed a deviation below 2 dB for a 1 kHz test tone compared to a calibrated sound level meter.

In this paper, we present a novel data-driven approach to quantify safety for non-linear, discrete-time stochastic systems with unknown noise distribution. We define safety as the probability that the system remains in a given region of the state space for a given time horizon and, to quantify it, we present an approach based on Stochastic Barrier Functions (SBFs). In particular, we introduce an inner approximation of the stochastic program to design an SBF in terms of a chance-constrained optimisation problem, which allows us to leverage the scenario approach theory to design an SBF from samples of the system with Probably Approximately Correct (PAC) guarantees. Our approach leads to tractable, robust linear programs, which enable us to assert safety for nonlinear models with a confidence that was otherwise deemed infeasible with existing methods. To further mitigate the computational complexity of our approach, we exploit the structure of the system dynamics and rely on spatial data structures to accelerate the construction and solution of the underlying optimisation problem. We show the efficacy and validity of our framework in several benchmarks, showing that our approach can obtain tighter certificates compared to state-of-the-art with a confidence that is several orders of magnitude higher.

Automated vehicles (AVs) consistently encounter ethically ambiguous situations in everyday driving, scenarios involving conflicting human interests and no clearly optimal course of action. While existing work often focuses on rare, high-stakes dilemmas (e.g., crash avoidance or trolley problems), routine decisions such as overtaking cyclists or navigating social interactions remain underexplored. This study addresses that gap by applying the tracking condition of Meaningful Human Control (MHC), which holds that AV behaviour should align with human reasons—the values, intentions, or expectations that justify actions. We conducted semi-structured interviews with 18 AV experts, who explained the reasons behind the considerations AV should make when planning a manoeuvre. Thirteen reason categories emerged, organised across normative, strategic, tactical, and operational levels. Using a case study on cyclist overtaking, we demonstrate how these reasons interact in practice and expose tensions in the decision-making process. Building on this analysis, we derive a reason-prioritisation principle grounded in the cyclist-overtaking scenario for AV behaviour in ethically ambiguous routine situations: prioritising vulnerable road users’ safety above all, treating systemic safety and regulation as important but conditional, and permitting secondary values only when safety is not compromised. This hierarchy supports human-aligned behaviour by allowing pragmatic actions when strict legal compliance would undermine higher-priority values. Our findings offer conceptual principles intended to inform future research and design for AV decision-making in ethically challenging routine situations.

Objective: This review systematically investigates the determination of sufficient time budgets in conditionally automated driving, focusing on the interplay between takeover time, allocated time budgets, and outcomes.

Background: Conditionally automated driving requires human drivers to resume vehicle control within limited time budgets when system limits are reached. However, the significant variability in drivers’ takeover time (the time needed to regain control) poses challenges in balancing time budgets to avoid being too short (compromising safety and comfort) or too long (reducing driver alertness). Prior work lacks systematic exploration of sufficient time budgets across scenarios and drivers.

Method: Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, 100 articles are selected for review. Review papers are analyzed to extract overarching insights on time budget design, while primary empirical studies are examined to complement and validate the findings. We systematically synthesize evidence along the takeover sequence, covering (i) drivers’ required takeover time, (ii) the time budgets provided by automated driving systems, and (iii) the resulting takeover outcomes. Based on the synthesis, we further discuss how the reviewed evidence informs the design of adaptive time budgets.

Results: This review synthesizes evidence on the takeover sequence and shows that takeover time varies significantly across drivers and driving contexts. Fixed time budgets cannot accommodate such variability. Adaptive time budgets that adjust to driver takeover demands are therefore promising. We propose an adaptive framework in which the time budget is defined as the sum of a predicted takeover time and an additional takeover buffer. To support this framework, the review identifies a qualitative inverted-U relationship among takeover time, time budget, and takeover outcomes: outcomes improve when the time budget exceeds the predicted takeover time, but deteriorate when the budget becomes unnecessarily long. Conclusion: Structuring the takeover sequence clarifies trade-offs in time budget design. Adaptive approaches that estimate takeover time and allocate an appropriate takeover buffer to achieve targeted outcomes show promise yet warrant further investigation. Application: Insights can optimize human-vehicle interactions for safe and comfortable takeovers, thus promoting public trust and acceptance of conditionally automated driving.

Partially automated driving systems are designed to perform specific driving tasks—such as steering, accelerating, and braking—while still requiring human drivers to monitor the environment and intervene when necessary. This shift of driving responsibilities from human drivers to automated systems raises concerns about accountability, particularly in scenarios involving unexpected events. To address these concerns, the concept of meaningful human control (MHC) has been proposed. MHC emphasises the importance of humans retaining oversight and responsibility for decisions made by automated systems. Despite extensive theoretical discussion of MHC in driving automation, there is limited empirical research on how real-world partially automated systems align with MHC principles. This study offers two main contributions: (1) an empirical evaluation of MHC in partially automated driving, based on 103 semi-structured interviews with users of Tesla's Autopilot and Full Self-Driving (FSD) Beta systems; and (2) a methodological framework for assessing MHC through qualitative interview data. We operationalise the previously proposed tracking and tracing conditions of MHC using a set of evaluation criteria to determine whether these systems support meaningful human control in practice. Our findings indicate that several factors influence the degree to which MHC is achieved. Failures in tracking—where drivers' expectations regarding system safety are not adequately met—arise from technological limitations, susceptibility to environmental conditions (e.g., adverse weather or inadequate infrastructure), and discrepancies between technical performance and user satisfaction. Tracing performance—the ability to clearly assign responsibility—is affected by inconsistent adherence to safety protocols, varying levels of driver confidence, and the specific driving mode in use (e.g., Autopilot versus FSD Beta). These findings contribute to ongoing efforts to design partially automated driving systems that more effectively support meaningful human control and promote more appropriate use of automation.

Identifying driving heterogeneity plays an important role in improving traffic safety and efficiency. This paper proposes a novel framework to identify driving heterogeneity from the underlying characteristics of driving behaviour. The framework includes three processes: Action phase extraction, Action pattern calibration, and Action pattern classification. The concepts of Action phase and Action patterns are proposed to decipher and interpret driving behaviours. Action phases are extracted by rule-based segmentation methods and Action patterns are calibrated based on an unsupervised learning approach. The extraction and calibration processes provide a rigorous labelling approach for the attention-based LSTM Action pattern classification process. Evaluation of the framework on a large-scale naturalistic driving dataset reveals six distinct Action patterns. The implementation of the attention mechanism to LSTM models significantly enhanced both the accuracy and time efficiency of Action pattern identification. The proposed framework offers benefits in detecting and reducing variability in driving behaviour through ITS applications such as user-based traffic management, personalised Advanced Driver Assistance Systems (ADAS), and advanced autonomous vehicles (AV) design, thereby enhancing road safety and traffic efficiency.

Driving heterogeneity significantly influences traffic performance, contributing to traffic disturbances, increased crash risks, and inefficient fuel use and emissions. With the growing availability of driving behaviour data, Machine Learning (ML) techniques have become widely used for analysing driving behaviour and identifying heterogeneity. This paper presents a systematic review of current ML-based methods for driving heterogeneity identification. The review organises key concepts and categorisations of driving heterogeneity, highlights strengths and drawbacks of various methods, and outlines applications of identification analysis. Based on the literature review, we propose a structured framework that guides the ML-based identification process. The framework starts with an extensive data collection and rigorous pre-processing process, followed by feature selection techniques that identify features most indicative of driving behaviours. Sophisticated models including supervised, unsupervised, semi-supervised, and reinforcement learning techniques are discussed with multi-perspective performance evaluation. This paper provides a comprehensive reference for researchers and practitioners to understand driving heterogeneity, supporting the development of data-driven solutions for improving traffic management and road safety.

Ethical dilemmas are a common challenge in everyday driving, requiring human drivers to balance competing priorities such as safety, efficiency, and rule compliance. However, much of the existing research in automated vehicles (AVs) has focused on high-stakes "trolley problems,"which involve extreme and rare situations. Such scenarios, though rich in ethical implications, are rarely applicable in real-world AV decision-making. In practice, when AVs confront everyday ethical dilemmas, they often appear to prioritise strict adherence to traffic rules. By contrast, human drivers may bend the rules in context-specific situations, using judgement informed by practical concerns such as safety and efficiency. According to the concept of meaningful human control, AVs should respond to human reasons, including those of drivers, vulnerable road users, and policymakers. This work introduces a novel human reasons-based supervision framework that detects when AV behaviour misaligns with expected human reasons to trigger trajectory reconsideration. The framework integrates with motion planning and control systems to support real-time adaptation, enabling decisions that better reflect safety, efficiency, and regulatory considerations. Simulation results demonstrate that this approach could help AVs respond more effectively to ethical challenges in dynamic driving environments by prompting replanning when the current trajectory fails to align with human reasons. These findings suggest that our approach offers a path toward more adaptable, human-centered decision-making in AVs.

The demand for air transport has experienced rapid growth, raising significant environmental concerns. Previous studies on airport sustainability have mainly focussed on airside areas; while literature pertaining to landside transport sustainability and emissions reduction approaches is limited. This paper addresses this gap by presenting a comprehensive review and critical assessment of the existing studies aiming to inform policymakers and researchers. Based on a holistic approach and an interdisciplinary perspective, the reviewed literature is classified into four categories: travel behaviour, transport infrastructure, transport policy, and sustainability. Within these categories, key findings are identified, along with a concise overview of the strategies employed by large airports to achieve their net-zero targets. Main solutions include increasing the attractiveness of public transport, providing electric vehicle (EV) infrastructure, and promoting low/zero emission vehicles. Moreover, the exploration of innovative fuel technologies (such as hydrogen), fast-charging EVs, and autonomous shuttle buses also has immense potential.

With a plethora of different seemingly diverging expansions for use of Meaningful Human Control (MHC) in practice, this paper proposes an alignment for the operationalisation of MHC for autonomous systems by proposing operational principles for MHC and introducing a generic framework for its application. The increasing integration of autonomous systems in various domains emphasises a critical need to maintain human control to ensure responsible safety, accountability, and ethical operation of these systems. The concept of MHC offers an ideal concept for the design and evaluation of human control over autonomous systems, while considering human and technology capabilities. Through conceptual synthesis of existing literature and investigation across various domains and related concepts, principles for the operationalisation of MHC are set out to provide tangible guidelines for researchers and practitioners aiming to implement MHC in their systems. The proposed framework dissects generic components of systems and their subsystems aligned with different agents, stakeholders and processes at different levels of proximity to an autonomous technology. The framework is domain-agnostic, emphasizing the universal applicability of the MHC principles irrespective of the technological context, paving the way for safer and more responsible autonomous systems.

BACKGROUND: Long-term noise annoyance can be expected to have worse outcomes than short-term annoyance. This study investigates noise annoyance over time, its association with personality traits and potential reciprocal effects between health outcomes and noise annoyance. METHODS: Firstly, we conducted a Longitudinal Latent Class Analysis to identify noise annoyance profiles. We further analysed the effect of Big Five personality traits on the likelihood of belonging to these annoyance profiles. Secondly, we used Cross-lagged Panel Models to analyse whether changes in noise annoyance precede changes in health outcomes or vice versa. For both analyses, we used 8 years of data from the Dutch Longitudinal Internet Studies for the Social Sciences (LISS) panel. Between 2708 and 11,068 subjects were included (this varies between models). RESULTS: We found three profiles of noise annoyance, namely, chronically, occasionally and never annoyed. Among all participants, 12% were chronically annoyed by neighbour noise and 6% by street noise. Extraversion and emotional stability decreased the chance of belonging to the cluster of chronically annoyed, while openness had the opposite effect. Chronic noise annoyance showed a significant effect on self-reported heart complaints and sleeping problems, while the effects of noise annoyance profiles on high blood pressure and heart attacks were insignificant. Some potential indications for a reverse effect from health outcomes on noise annoyance were found. CONCLUSION: Noise annoyance was relatively stable over time possibly because of its correlation with personality traits. Noise had a small negative effect on health outcomes, and some health outcomes affected noise annoyance. Further research should be conducted to collect dedicated panel data.