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.

Collision avoidance – involving a rapid threat detection and quick execution of the appropriate evasive maneuver – is a critical aspect of driving. However, existing models of human collision avoidance behavior are fragmented, focusing on specific scenarios or only describing certain aspects of the avoidance behavior, such as response times. This paper addresses these gaps by proposing a computational cognitive model of human collision avoidance behavior based on active inference. Active inference provides a unified approach to modeling human behavior: the minimization of free energy. Building on prior active inference work, our model incorporates established cognitive mechanisms such as evidence accumulation to simulate human responses in three distinct collision avoidance scenarios: front-to-rear lead vehicle braking, lateral incursion by an oncoming vehicle, and another vehicle failing to yield at an intersection. We demonstrate that our model explains a wide range of empirical findings on human collision avoidance behavior. Specifically, the model closely reproduces both aggregate results from meta-analyses previously reported in the literature and detailed, scenario-specific effects observed in two recent driving simulator studies, including response timing, maneuver selection, and execution. Our results highlight the potential of active inference as a generalizable framework for understanding and modeling human behavior in complex real-life driving tasks.

Merging onto a highway is a safety-critical task resulting in a large number of traffic accidents; fundamental research into merging behavior of human drivers can help reduce this toll. Two cognitive processes critical to merging, attention allocation and decision making, have been extensively studied in real-world and simulated driving scenarios. However, how these processes interact during highway merging remains poorly understood. While the relationship between attention and decision making has been widely examined in cognitive science, this work has largely relied on simple decision-making paradigms involving choices between static items on a computer screen, which limits the understanding of more dynamic and naturalistic decisions such as in driving. To address this gap, we investigated the relationship between attention and decision making in a simplified highway merging task. In a video-based experiment, participants (N=24) repeatedly made merging gap acceptance decisions based on the dynamic information about the distance and time-to-arrival to the end of the merging lane and the gap to the target-lane vehicle (available in the front view and the side mirror, respectively). Participants’ decisions, response times, and eye movements were recorded. We found that decisions to accept a gap were considerably faster than decisions to reject a gap. Decision outcomes and timing depended on the distance to and time-to-arrival of the target-lane vehicle, but also on the time pressure due to approaching the end of the merging lane. Most importantly, under high time pressure, a greater proportion of time spent looking at the side mirror was associated with a lower probability of accepting the gap. This finding indicates that differences in visual information sampling can be closely linked to decision outcomes when time budgets are constrained. Our results provide initial empirical insights relevant for future cognitive modeling of the interplay between decision making and attention during highway merging. This work can inform early-stage exploration of driver monitoring and support systems for partially automated driving.

Upper extremity (UE) impairment is a common consequence of stroke, restricting daily activities. Clinical assessments such as the Fugl–Meyer Assessment (FMA) and the Action Research Arm Test (ARAT) are widely used but are typically therapist-administered. Inertial measurement units (IMUs) provide a portable, objective method to quantify upper limb kinematics and may therefore support scalable tele-rehabilitation. Yet, evidence on their reliability, validity, and clinical relevance remains limited. This study evaluated the test–retest reliability, discriminant validity (vs. healthy controls), and convergent validity (correlation with FMA and ARAT) of eleven IMU-derived kinematic metrics during a standardized drinking task in individuals with subacute stroke. Fifteen stroke patients and fifteen healthy controls performed the task wearing four IMUs on the upper limb and sternum. Both joint and end-point kinematics were derived using the Madgwick sensor fusion algorithm. Reliability was assessed through intraclass correlation coefficients (ICCs), discriminant validity through linear mixed models (LMMs), and convergent validity through Pearson’s correlations and regression models. Most metrics showed good to excellent reliability (ICC≥0.75), except for shoulder abduction (ICC=0.18) and maximum elbow angular velocity (ICC=0.65). All but shoulder abduction demonstrated significant discriminant validity. Movement time and measures of smoothness correlated moderately to strongly (r≥.67) with ARAT and FMA. These findings indicate that IMU-derived metrics during a standardized drinking task provide reliable, valid, and clinically meaningful insights into post-stroke motor status, and may offer supplementary information for movement assessment beyond conventional clinical scales.

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.

Light detection and ranging (LiDAR) plays a crucial role in machine perception for advanced driver assistance systems. Existing LiDARs, however, do not adapt their sensing strategy to complement driver's perception. We demonstrate a novel LiDAR prototype that dynamically adapts its range and resolution over the field of view, according to real-time driver gaze. Our gaze-aware LiDAR emphasizes scanning peripheral zones the driver may overlook, i.e., critical areas during driving. Our demonstration showcases enhanced perception, highlighting the potential of hybrid human-machine sensing for safer driving.

One of the key issues in human-robot collaboration is the development of computational models that allow robots to predict and adapt to human behavior. Much progress has been achieved in developing such models, as well as control techniques that address the autonomy problems of motion planning and decision-making in robotics. However, the integration of computational models of human behavior with such control techniques still poses a major challenge, resulting in a bottleneck for efficient collaborative human-robot teams. In this context, we present a novel architecture for human-robot collaboration: Adaptive Robot Motion for Collaboration with Humans using Adversarial Inverse Reinforcement learning (ARMCHAIR). Our solution leverages adversarial inverse reinforcement learning and model predictive control to compute optimal trajectories and decisions for a mobile multi-robot system that collaborates with a human in an exploration task. During the mission, ARMCHAIR operates without human intervention, autonomously identifying the necessity to support and acting accordingly. Our approach also explicitly addresses the network connectivity requirement of the human-robot team. Extensive simulation-based evaluations demonstrate that ARMCHAIR allows a group of robots to safely support a simulated human in an exploration scenario, preventing collisions and network disconnections, and improving the overall performance of the task.

Background: Robotic devices have shown promise in supporting motor (re)learning. However, there is a limited understanding of how personality traits influence the effectiveness of robot-aided training strategies. Methods: We conducted a motor learning experiment with 40 unimpaired participants who trained to control a virtual pendulum using a robotic haptic device. Before the experiment, we assessed personality traits including the perceived control over life events (Locus of Control), the tendency to turn challenges into engaging activities (Transform of Challenge), and other subscales from Autotelic and Hexad gaming style questionnaires. Participants were divided into two groups, one receiving haptic guidance during training and a second one without assistance. Short- and long-term retention was assessed, and relationships between personality traits, performance metrics, and human-robot interaction metrics were analyzed. Results: Participants with high Transform of Challenge or external Locus of Control characteristics who received physical guidance during training reduced the human-robot interaction forces to a lesser extent compared to the ones who did not receive guidance. Additionally, participants with a high Free Spirit gaming style showed greater sensitivity to how their perception of the guidance affected their performance during the retention phases. Conclusion: Our findings suggest that autotelic personality, Locus of Control, and gaming style modulate motor learning outcomes during robotic-assisted training, affecting both performance and human-robot interaction metrics. This highlights the potential of integrating personality-based adaptations in robot-aided rehabilitation protocols to enhance performance and motor (re)learning. Future works should explore the relationship between personality traits and psychological states (e.g., perceived difficulty, attention) across diverse tasks and guidance methods in clinical populations.

Home-based rehabilitation is essential for stroke survivors, facilitating motor recovery and improving activities-of-daily-life performance. Recent advances in wearable technologies and machine learning promise to revolutionize home-based arm rehabilitation by providing detailed movement analysis. However, machine learning algorithms for arm movement identification are predominantly trained and tested in the same environments. Their ability to generalize to novel environments remains largely unknown, hindering practical applications. This paper investigates the ability of two established machine learning models to generalize a structured, lab-based environment to a more realistic, semi-structured kitchen environment. Twelve healthy participants performed various arm activities, involving three arm movement types (reaching, lifting, and pronation/supination). In addition to evaluating the generalization of movement identification, we compared algorithm performance for two different sensor configurations: four Inertial Measurement Units (IMUs) on the arm versus a single IMU on the wrist. We employed a Random Forest (RF) classifier and a hybrid deep learning model combining convolutional and recurrent neural networks, evaluating both subject-specific and group approaches. Trained in the structured environment, the RF classifier predicted activities in the semi-structured environment with 86.54% (subject-specific) and 77.37% (group) balanced accuracy, based on the four-sensor configuration, while the hybrid model reached 87.96% and 82.96% accuracy. The accuracy was lower with a single wrist IMU; the RF classifier showed a smaller decrease than the hybrid model. Our findings demonstrate that the investigated arm movement identification algorithms generalize well across environments even with the minimal sensor configuration, indicating the potential for future applications in home-based stroke rehabilitation.

Ensuring safe interactions between autonomous vehicles (AVs) and human drivers in mixed traffic systems remains a major challenge, particularly in complex, high-risk scenarios. This paper presents a cognition-decision framework that integrates individual variability and commonalities in driver behavior to quantify risk cognition and model dynamic decision-making. First, a risk sensitivity model based on a multivariate Gaussian distribution is developed to characterize individual differences in risk cognition. Then, a cognitive decision-making model based on the drift diffusion model (DDM) is introduced to capture common decision-making mechanisms in high-risk environments. The DDM dynamically adjusts decision thresholds by integrating initial bias, drift rate, and boundary parameters, adapting to variations in speed, relative distance, and risk sensitivity to reflect diverse driving styles and risk preferences. By simulating high-risk scenarios with lateral, longitudinal, and multidimensional risk sources in a driving simulator, the proposed model accurately predicts cognitive responses and decision behaviors during emergency maneuvers. Specifically, by incorporating driver-specific risk sensitivity, the model enables dynamic adjustments of key DDM parameters, allowing for personalized decision-making representations in diverse scenarios. Comparative analysis with IDM, Gipps, and MOBIL demonstrates that DDM more precisely captures human cognitive processes and adaptive decision-making in high-risk scenarios. These findings provide a theoretical basis for modeling human driving behavior and offer critical insights for enhancing AV-human interaction in real-world traffic environments.

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.

Safe and socially acceptable interactions with human-driven vehicles are a major challenge in automated driving. A good understanding of the underlying principles of such traffic interactions could help address this challenge. Particularly, accurate driver models could be used to inform automated vehicles in interactions. These interactions entail complex dynamic joint behaviors composed of individual driver contributions in terms of high-level decisions, safety margins, and low-level control inputs. Existing driver models typically focus on one of these aspects, limiting our understanding of the underlying principles of traffic interactions. Here, we present a Communication-Enabled Interaction model based on risk perception, that does not assume humans are rational and explicitly accounts for communication between drivers. Our model can explain and reproduce observed human interactions in a simplified merging scenario on all three levels. Thereby improving our understanding of the underlying mechanisms of human traffic interactions and posing a step towards interaction-aware automated driving.

The use of partially automated driving systems raises concerns about potential responsibility issues, posing risk to the system safety, acceptance, and adoption of these technologies. The concept of meaningful human control has emerged in response to the responsibility gap problem, requiring the fulfillment of two conditions, tracking and tracing. While this concept has provided important philosophical and design insights on automated driving systems, there is currently little knowledge on how meaningful human control relates to subjective experiences of actual users of these systems. To address this gap, our study aimed to investigate the alignment between the degree of meaningful human control and drivers' perceptions of safety and trust in a real-world partially automated driving system. We utilized previously collected data from interviews with Tesla "Full Self-Driving"(FSD) Beta users, investigating the alignment between the user perception and how well the system was tracking the users' reasons. We found that tracking of users' reasons for driving tasks (such as safe maneuvers) correlated with perceived safety and trust, albeit with notable exceptions. Surprisingly, failure to track lane changing and braking reasons was not necessarily associated with negative perceptions of safety. However, the failure of the system to track expected maneuvers in dangerous situations always resulted in low trust and perceived lack of safety. Overall, our analyses highlight alignment points but also possible discrepancies between perceived safety and trust on the one hand, and meaningful human control on the other hand. Our results can help the developers of automated driving technology to design systems under meaningful human control and are perceived as safe and trustworthy.

Traffic jams occurring on highways cause increased travel time as well as increased fuel consumption and collisions. So-called phantom traffic jams are traffic jams that do not have a clear cause, such as a merging on-ramp or an accident. Phantom traffic jams make up 50% of all traffic jams and result from instabilities in the traffic flow that are caused by human driving behavior. Automating the longitudinal vehicle motion of only 5% of all cars in the flow can dissipate phantom traffic jams. However, driving automation introduces safety issues when human drivers need to take over the control from the automation. We investigated whether phantom traffic jams can be dissolved using haptic shared control. This keeps humans in the loop and thus bypasses the problem of humans’ limited capacity to take over control, while benefiting from most advantages of automation. In an experiment with 24 participants in a driving simulator, we tested the effect of haptic shared control on the dynamics of traffic flow and compared it with manual control and full automation. We also investigated the effect of two control types on participants’ behavior during simulated silent automation failures. Results show that haptic shared control can help dissipating phantom traffic jams better than fully manual control but worse than full automation. We also found that haptic shared control reduces the occurrence of unsafe situations caused by silent automation failures compared to full automation. Our results suggest that haptic shared control can dissipate phantom traffic jams while preventing safety risks associated with full automation.