Older drivers are among the most vulnerable demographics within the road traffic system. The rising number of elderly motorists has raised public concern regarding their driving safety. It is crucial to understand the factors influencing risky driving behaviors among older drivers to enhance their safety. This study aimed to analyze the personality, functional performance, and travel patterns related to older drivers' risky driving behavior. The analysis utilized a sample of 58 older drivers, aged 65 years and above (mean age = 72.41 years; 40 males and 18 females) from the Nagoya metropolitan area. Risky driving behaviors and travel patterns were assessed using naturalistic driving data. Bivariate correlation analysis revealed that impulsivity and diminished contrast sensitivity were significantly correlated with more frequent risky driving behaviors. Additionally, both low driving exposure and high-risk driving routes (i.e., more frequent left and right turns, driving more on minor roads) were significantly correlated with an increased risk of harsh events. Moreover, a strong association was observed between driving exposure and driving route, indicating that the driving route of lower mileage drivers tend to be riskier. When the relationship between driving exposure and risky driving behaviors was adjusted for driving route, the strength of the correlation diminished from 0.35 to 0.16, rendering it insignificant. This partial correlation analysis suggests that the increased driving risk among low-mileage drivers can be partially attributed to their high-risk driving routes. The findings of this study provide further evidence regarding the role of personality in explaining older drivers' risky driving behavior and the explanation of older drivers' low-mileage bias.

The lane-changing (LC) maneuver of vehicles significantly impacts highway traffic safety. Therefore, proactively predicting LC maneuver and associated collision risk is of paramount importance. However, most of the previous LC risk prediction research overlooks the prediction of LC maneuver, limiting its practical utility. Furthermore, the effectiveness of LC maneuver recognition tends to be moderate as the prediction horizon extends. To fill the gaps, this paper proposes a multi-task learning model that simultaneously predicts the probability of LC maneuver, LC risk level, and time-to-lane-change (TTLC), while further analyzing the intrinsic correlation between LC maneuver and LC risk. The model consists of a Convolutional Neural Network (CNN) and two Long Short-Term Memory networks (LSTM). The CNN is employed to extract and fuse shared features from the dynamic driving environment, while one LSTM is dedicated to estimating the probability of LC maneuver and TTLC, and the other LSTM focuses on estimating the LC risk level. Evaluation of the proposed method on the HighD dataset demonstrates its excellent performance. It can almost predict all LC maneuvers within 2 s before the vehicle crosses lane boundaries, with an 80% recall rate for high-risk LC levels. Even 3.6 s before crossing lane boundaries, the model can still predict approximately 95% of LC maneuvers. The use of the multi-task learning strategy enhances the model's understanding of traffic scenarios and its prediction robustness. LC risk analysis based on the HighD dataset shows that the risk distribution and influencing factors for left and right lane changes differ. In right lane changes, collision risks primarily arise from the leading and following vehicles in the current lane, while in left lane changes, collision risks mainly stem from the leading vehicle in the current lane and the following vehicle in the target lane. The proposed approach can be applied to advanced driver assistance systems (ADAS) to reliably and early identify LC during highway driving, while correcting potentially dangerous LC maneuvers, ensuring driving safety.

The eight-lane freeway resulting from reconstruction and expansion typically exhibits short distances between interchanges and a wide road section. Nonetheless, the absence of specific guidelines for the placement of guide signs in dense interchange sections of the eight-lane freeway results in inadequate design, thereby impeding drivers' ability to read and comprehend the signs. To tackle this issue, the study employs two interchanges 2.48 km apart on the Jinan-Qingdao Freeway as a case study. Four optimization schemes for guide signs are developed based on drivers' information requirements and compared with the current guide sign design scheme. Thirty-nine drivers were recruited to gather detailed driving behavior indicators via a driving simulation experiment. The impact of the guide sign optimization scheme on driving behavior is analyzed, and the overall effects are evaluated using the non-integer rank RSR method. This study aims to identify an optimal approach to guide sign design for dense interchange sections. The results indicate that the impact of guide signs in dense interchange sections on drivers is primarily concentrated between the two interchanges. Specifically, the addition of a 2.5 km exit advance sign enhances drivers' speed regulation level, the inclusion of navigation voice improves operational stability, and the presence of pavement words at exit diversion locations enhances psychological comfort for drivers. By considering the comprehensive effectiveness of each optimization scheme, it is evident that schemes 5 and 2 exhibit superior optimization effects. This suggests that providing advanced notice of exit information in dense interchange sections of eight-lane freeways is an effective measure to enhance freeway service levels and ensure driving safety. It is recommended that under the conditions of insufficient interchange spacing, the information of interchange exits should be forewarned in advance. Additionally, auxiliary navigation voice and pavement words should be employed to enhance drivers' information perception levels, thereby mitigating the risk of missing exits due to limited reaction time. This paper serves as a significant reference for informing the optimal configuration of guide signs, thereby contributing to the meticulous development of standardized specifications.

One of the major challenges in automated driving is ensuring that the system can handle all possible driving scenarios, including rare and critical ones, also referred to as corner case scenarios. For the validation of automated driving functions, it is necessary to test the corner cases in simulation environments. However, the effectiveness of simulation-based testing depends on the availability of realistic test data that accurately reflect real-world scenarios. This work aims to detect, cluster, and analyze rare and critical traffic scenarios based on real-world traffic data from an urban intersection and prepare the data for usage in simulation environments. The scenarios are detected by filtering hard braking maneuvers, red light violations, and near misses under adverse weather conditions. A long-term analysis of trajectory, weather, and traffic light data was conducted to find these rare scenarios. Our results show that 24 hard braking maneuvers are included in our dataset with a duration of half a year. They occur due to failure to yield, emergency vehicle operations, and a red light violation. Some of the scenarios include crashes, lateral evasive maneuvers, or are under adverse weather conditions like fog. Altogether, we provide methods to extract corner case scenarios based on multiple data sources and reveal diverse types of corner case scenarios at an urban intersection. In addition, we analyze the behavior of road users in critical scenarios and show influencing factors to avoid crashes. By combining and converting the data to an industry standard for simulation we provide realistic test cases for the validation of automated vehicles. Therefore, the results are relevant for both, traffic safety researchers to learn from road user behavior in these rare scenarios and developers of automated driving systems to test their functions.

A comprehensive investigation into the mechanisms and causes of traffic crashes holds significant implications for crash prevention and mitigating crash injury severity. Under the influence of unobservable factors, the impact of the same factor on crash injury severity might not only vary spatially but also exhibit temporal instability. Neglecting these characteristics could lead to biased model estimations and confounding effects, potentially resulting in ineffective or even counterproductive traffic safety strategies. Simultaneously considering the spatial heterogeneity and temporal instability of factors that influence crash injury severity, this paper first collects traffic crash data from the Austin metropolitan area in Texas, USA, spanning the years 2017 to 2019, where various independent variables are selected as candidate variables for analyzing crash injury severity, and a latent class logit model is constructed. Subsequently, annual traffic-related statistical exogenous data involving 11 counties are utilized to establish class probability functions within the latent class logit model, thereby accounting for the spatial heterogeneity of crash injury severity. Finally, this study conducts the partially constrained approach for modeling annual basis, simultaneously analyzing the temporal instability of safety factors' impact on crash injury severity. Notably, this paper not only identifies numerous factors significantly influencing crash injury severity but also discovers that certain factors exhibit significant temporal instability effects on crash injury severity. Several explanatory variables showed temporally instability in terms of their effect on resulting injury severities. Such as, crash locations, lighting conditions, driver age, driver gender, vehicle types, vehicle model year. The findings of this study serve as a valuable reference for delving deeper into the causal mechanisms of crash injury severity as well as formulating effective safety measures.

In recent years, micromobility has seen unprecedented growth, especially with the introduction of dockless e-scooters. However, the rapid emergence of e-scooters has led to an increase in crashes, resulting in injuries and fatalities, highlighting the need for in-depth analysis to understand the underlying mechanisms. While helpful in quantifying the problem, traditional crash database analysis cannot fully explain the causation mechanisms, e.g., human adaptation failures leading to safety-critical events. Naturalistic data have proven extremely valuable for understanding why crashes happen, but most studies have addressed cars and trucks. This study is the first to systematically analyze factors contributing to crashes and near-crashes involving rental e-scooters in an urban environment, utilizing naturalistic data. The collected dataset included 6868 trips, covering 9930 km over 709 h with 4694 unique participants. We identified 61 safety-critical events, including 19 crashes and 42 near-crashes, and subsequently labeled variables associated with each event according to the codebook using video data. Our odds ratio analysis identified that rider experience and behavior (e.g., phone usage, single-handed riding, and pack riding) significantly increase the crash risk. Given the accessibility of rental e-scooters to individuals regardless of their experience, our findings emphasize the need for rider training in addition to education. Influenced by their experience with bicycles, riders may anticipate a similar self-stabilizing mechanism in e-scooters. We found that single-handed riding, which compromises balance, poses a heightened risk, underscoring the crucial role of balance in safe e-scooter operation. Furthermore, the purpose (leisure or commute) and directness (point-to-point or detour) of the trip were also identified as factors influencing the risk, suggesting that user intent plays a role in safety-critical events. Interestingly, our analysis underscores the importance of adapting the crash and near-crash definitions when working with two-wheeled vehicles, especially those in the shared mobility system.

Bicycle crashes at intersection areas are posed a worrying traffic safety issue, and one of the main reasons for bicycle crashes is failing to avoid conflicts with motor vehicles and other bicycles. Clearly, cyclists are more exposed to risk if they perform a direct left turn (DLT) being mixed with left-turning vehicle under a left-turn phase. Owing to the lack of exposure data, the detection of DLT event and the mechanism behind the risky riding behavior have yet to be discovered. To bridge these gaps, this study proposes a DLT detection framework based on bike sharing trajectories. Moreover, this study seeks to understand the contributing factors to DLT behavior using the random parameters logit model with heterogeneity in means and variances (RPLHMV) to account for unobserved heterogeneity in the DLT cases dataset. Statistical analysis shows that DLT is most likely to occur on weekdays during peak periods under large commuting demand. As to what caused the DLT violations, law-obeying cyclists are more susceptible to external events, while risk-taking cyclists are subtly undermined by their habits. In addition, the model of RPLHMV reveals several significant contributing factors to the propensity of DLT violations, such as event time, available passing time for left-turning bicycles, and average cycling speed, whereas the indicator variables of actual waiting time, available passing space for left-turning bicycles, and preference for DLT violation become the emerging influential variables. This study is expected to help better understand DLT occurrence and propose countermeasures more efficiently for reducing cyclists' DLT rate.

This paper investigates the impacts of winter maintenance operations (WMO) on road safety under different weather conditions using connected vehicle data. In particular, the impacts of WMO on incident-induced delays (IID) and harsh braking events are highlighted, representing the influence on traffic flow and vehicle stability, respectively. Taking advantage of emerging connected vehicle data, the impacts of WMO on IIDs and vehicle harsh braking events are estimated. Data analysis revealed that WMO plays an important role in reducing the mean IID and the average number of harsh braking events, particularly when roads were covered with ice, frost, slush, or snow in snowy weather. The presence of WMO reduced the mean IID from 145.93 veh-h to 57.70 veh-h, representing a 60% decrease, and the number of harsh braking events from 3.58 cases per crash to 2.90 cases per crash, making a 19% reduction. Last, the multiple linear regression (MLR) model highlights that WMO effectively reduces IID by 23.36 veh-h. In addition, the MLR model indicates that IID is influenced by traffic volume, driving behaviors immediately before a crash, crash severity, road weather conditions, with more severe crashes and worse pavement conditions contributing to longer delays. These findings suggest that the WMO can improve road safety by reducing incident-induced delays and improving traffic stability in winter weather conditions.

Jaywalking, as a hazardous crossing behavior, leaves little time for drivers to anticipate and respond promptly, resulting in high crossing risks. The prevalence of Autonomous Vehicle (AV) technologies has offered new solutions for mitigating jaywalking risks. In this study, we propose a risk-aware deep reinforcement learning (DRL) approach for AVs to make decisions safely and efficiently in jaywalker-vehicle interactions. Notably, a risk prediction module is incorporated into the traditional DRL framework, making the AV agent risk-aware. Considering the complexity of jaywalker-vehicle conflicts, an encoder-decoder model is adopted as the risk prediction module, which comprehensively integrates multi-source data and predicts probabilities of the final conflict severity levels. The risk-aware DRL approach is applied in a simulated environment established in Anylogic, where the motion features of jaywalkers and vehicles are calibrated using real-world survey data. The trained driving policies are evaluated from perspectives of safety and efficiency across three scenarios with escalading levels of jaywalker volume. Regarding safety performance, the Baseline policy performs the worst in "medium jaywalker volume" scenario and "high jaywalker volume" scenario, while our Proposed risk-aware method outperforms the other methods, with the "low TTC ratio" metric stabilizing near 0.08. Moreover, as the scenario gets more complex, the superiority of our Proposed risk-aware policy gets more evident. In terms of efficiency performance, our Proposed risk-aware policy ranks the second best, achieving an "AV delay" metric around 8.1 s in the "medium jaywalker volume" scenario and 8.5 s in the "high jaywalker volume" scenario. In practice, the proposed risk-aware DRL approach can help AV agents perceive potential risks in advance and navigate through potential jaywalking areas safely and efficiently, further enhancing pedestrian safety.

Safety is a critical aspect of traffic systems. However, traditional crash data-based methods suffer from scalability and generalization issues. Although SSMs offer a proactive alternative for safety evaluation, their validation in simulated settings remains inconsistent, especially with emerging mobility technologies like autonomous driving. Our study critiques existing methodologies in SSM validation and introduces a novel framework integrating micro-level driver models with macro-level traffic states. This approach accounts for diverse external factors, including weather and geographical variations. Utilizing the Caltrans Performance Measurement System (PeMS) data, we conduct a large-scale analysis, merging traffic simulation with real-world data to extract SSMs and correlate them with crash statistics. Our results indicate a significant correlation between SSM counts and crash numbers but no clear trend with varying SSM thresholds. This suggests limitations in current public data for establishing robust links between simulated SSMs and real-world crashes. Our study highlights the need for improved data collection and simulation techniques, paving the way for more accurate and meaningful roadway safety analysis in the era of advanced mobility systems.

Static obstacles (tunnel sidewalls, barricades, etc.) on the side of mountainous highways change the spatial range of the road during driving, restricting the driver's freedom of driving while possibly triggering the driver's shy away effect, which poses a specific potential safety hazard. To understand the characteristics of driving behaviour in mountain highway tunnels with different tunnel lengths and lateral obstacles, nine tunnels in Chongqing were selected for real-vehicle tests, and data on driving trajectories, speeds and other metrics were collected from 40 drivers. Analyse the driver's need for lateral safety distance in different scenarios, defines the conditions and scope of the shy away effect, and establishes a multi-scenario "distance-trajectory" offset prediction model to adjust the offset under varying lateral environments by setting different facilities. The results show that drivers exhibit some avoidance behavior towards lateral static obstacles, but the extent of the shy-away effect varies based on tunnel length. By widening the lateral clearance to 0.925 m on the left side and 1.450 m on the right side of the road to meet the driver's requirements for lateral safety distances, unreasonable avoidance behaviour can be reduced. Combined with the trajectory fluctuation characteristics of drivers in different tunnels, it is proposed to set up the traffic safety facilities in a manner more aligned with driver behavioral habits, with a place set up 110 m before the entrance of the short tunnel, two places set up in the medium tunnel at L/2 - 200 m, L/2 + 100 m (where L is the length of the tunnel), and three places for long tunnels at L/2 - 400 m, L/2 m, and L/2 + 300 m. For extra-long tunnels, facilities are to be set up in cycles of 500 m, 1000 m, and 1500 m intervals. In the cross-section where different drivers are prone to apparent trajectory offsets, a driving behavior prompt sign is added to help correct the driving trajectory.

Autonomous vehicles (AVs) should prioritise pedestrian safety in a traffic accident. External human-machine interfaces (eHMIs), which enhance communication through visual and auditory signals, become essential as AVs become prevalent. This study aimed to investigate the current state of research on eHMIs, with a specific focus on pedestrian interactions with eHMI-equipped AVs. A bibliometric analysis of 234 papers published between January 2014 and December 2023 was conducted using the Web of Science database. The analysis revealed a remarkable increase in eHMI research since 2018, with the principal research topics on crossing behaviour and eHMI evaluations of pedestrians. Subsequently, 38 articles were selected for a systematic review. The systematic review, conducted through a detailed examination of each selected article, showed that pedestrian crossing behaviour is usually measured using crossing initiation time, response time, walking speed and eye tracking data. The eHMI evaluations of pedestrians were made through questionnaires that measure clarity, preference and acceptance. Research findings showed that pedestrians' crossing behaviour and eHMI evaluations are influenced by human factors (age and nationality), vehicle factors (eHMI type, eHMI colour and eHMI position) and environmental factors (signalisation and distractions). The results also revealed that current eHMI experiments often use virtual reality and video methodologies, which do not fully replicate the complexities of real-world environments. Additionally, the exploration regarding the impact of human factors, such as gender and familiarity with AVs, on pedestrian crossing behaviour is lacking. Furthermore, the investigation of multimodal eHMI systems is limited. This review highlighted the importance of standardising eHMI design, and the key gaps in the current eHMI research were revealed. These insights will guide future research towards effective eHMI solutions through informed theoretical studies and practical applications in autonomous driving.

Driving stress is a critical factor leading to road traffic accidents. Despite numerous studies that have been conducted on driving stress recognition, most of them only focus on accuracy improvement without taking model interpretability into account. In this study, an explainable driving stress recognition framework was presented to quantify stress based on electroencephalography (EEG) and behavior data. Based on the extraction of key EEG and behavior features and feature selection, low, medium, and high levels of driving stress were identified using seven machine learning algorithms. The recognition results when only using EEG or behavior features were compared with the result when fusing EEG together with behavior features. Then, the dependency effects between brain activity, driving behavior, and stress were analyzed using the SHapley Additive exPlanation (SHAP) method, and fuzzy rules were obtained by decision tree method. Results indicated that after feature selection, the accuracy of the combined EEG and behavior feature set improved by 8.56% and 26.51% compared to the single EEG and behavior feature sets respectively, and the accuracy rate of 84.93% was achieved. Furthermore, the variations in driver behavior and physiology under stress were identified by the visualization results of SHAP and the quantitative analysis method of decision tree. The changes of different brain regions in the same frequency band showed higher synchronicity under driving stress stimulation. The changes caused by increased stress can be explained by lower speed, smaller maximum lateral lane deviation, smaller accelerator pedal depth and larger brake depth, along with the power changes of the θ and β-band of the brain.

The implementation of advanced driver assistance systems (ADAS) has significantly impacted the prevention of traffic accidents, particularly through the forward collision warning (FCW) algorithm. Nevertheless, traffic conflicts on traffic routes remain a significant issue, since most FCW algorithms cannot accurately determine the distance between the host vehicle (HV) and remote vehicle (RV) on curved roads. Hence, this study proposes a vector-based FCW (V-FCW) algorithm to address the issue of false warnings on unconventional road sections. The V-FCW algorithm employs vector relationships to estimate the poses of HV and RV at the current and next moments, thereby effectively calculating the relative angles. Firstly, the HV and RV transmit their position vector, velocity vector, and heading angle in real time via the vehicle-to-vehicle (V2V) communication technique. Subsequently, the localization of lanes is conducted through the vehicle-to-infrastructure (V2I) communication technique, with the assistance of roadside unit (RSU)-based local maps. Finally, a V-FCW algorithm was implemented on the Simcenter Prescan simulation platform and a cellular vehicle-to-everything (C-V2X, i.e., the combination of V2V and V2I) communication platform. The simulation results demonstrate that the proposed V-FCW algorithm can accurately identify and warn dangerous vehicles on both straight and curved roads. Moreover, the experimental results obtained from the hardware-in-the-loop approach illustrate the efficacy of the proposed V-FCW algorithm in accurately forecasting four warning levels on both straight and curved roads. Consequently, this study yields a significant contribution to the field of vehicle-road cooperation in C-V2X-enable intelligent driving.

Violations of road traffic law are widespread in all countries. Probably the most common violation is speeding. It is not uncommon that 50 % of vehicles are speeding. Little is known about the risk of apprehension for various traffic law violations, although it is often assumed that nearly all violations go undetected. This paper quantifies the risk of apprehension for common traffic law violations in Norway, based on data for the period 2006-2022. The violations included are speeding, non-use of seat belts, driving with an illegal blood alcohol concentration (above 0.02 %), driving while impaired by medicines or illegal drugs, use of a hand-held mobile phone while driving and violations of the regulations of hours of service and rest for drivers of heavy vehicles. Risk of apprehension is stated as the number of detected violations per million vehicle kilometres driven while committing the violation. The risk of apprehension is in most cases between 10 and 50 per million vehicle kilometres driven while committing a violation. This is quite low. For speeding, the risk of apprehension was between 10 and 12 per million vehicle kilometres of speeding during 2006-2022. For an average driver, this means that he or she could speed on every trip for about 8-10 years before getting caught. Reducing traffic law violations may contribute to a large reduction of the number of traffic fatalities.

Inadequate visibility is a critical factor contributing to the heightened occurrence of nighttime accidents at highway intersections. The installation of smart streetlights which are equipped to detect vehicle positions and speed information, thereby dynamically adjusting illumination, offers a promising solution to significantly reduce nighttime accident rates while conserving lighting energy. Nevertheless, as vehicles travel through illuminated intersections in a relative high speed and enter unlighted highway segments, drivers often experience dynamic visual illusions during dark adaptation, consequently impairing their stress response capacity and generating driving safety concerns. Therefore, we investigate the collaborative impact of illumination gradient and vehicle speed at intersection exits on driver stress response, aiming to provide a theoretical foundation for gradual illumination designs dynamically aligning with various vehicle speeds. Specifically, with reaction time employed as a metric to quantify driver stress response, and intersection area illuminance and vehicle speed utilized as input parameters, a safety assessment method for illumination gradients at exit sections is developed using variance analysis and multiple comparison techniques. Subsequently, a high-fidelity nighttime driving simulation platform is established, integrating initial illuminance, vehicle speed, and illumination gradient distance within exit sections as influential factors. Through simulated driving experiments, the collaborative effects of illumination gradient schemes and vehicle speed on reaction time is systematically examined. Ultimately, we propose optimal illumination gradient schemes and the minimum required number of streetlights for exit sections corresponding to specific vehicle speeds. Results reveal that exit section illumination is unnecessary when the vehicle speed is below 40 km·h. For vehicle speeds of 50, 60, and 70 km·h, the minimum required exit section lengths are determined to be 35, 70, and 105 m, respectively. Moreover, it is established that a minimum of one streetlight is indispensable within the exit section at a speed limit of 50 km·h, while at 60 km·h, at least two streetlights are required. Lastly, under a speed limit of 70 km·h, the exit section should accommodate no fewer than three streetlights to ensure optimal safety conditions.

The number of accidents involving elderly pedestrians has been increasing from year to year, in spite of various road safety initiatives having been implemented. In line with Singapore's ageing population, this presents a worrying trend. This study aims to shed light on possible contributing factors via a human factors analysis. A preliminary investigation was first conducted at traffic junctions identified to have a greater occurrence of accidents involving elderly pedestrians and motorists. This preliminary investigation looked into the efficacy of infrastructure-oriented solutions in reducing the occurrence of such accidents. It was observed that infrastructure alone was inadequate in ensuring safety of elderly pedestrians. Next, a questionnaire was administered in order to gain information regarding traits, attitudes and behaviours pertinent to traffic safety. Subsequently, structural equation modelling was used to analyse the data via exploratory, confirmatory and path analysis. This was followed by an in-depth discussion which explored the relationship between the latent constructs of traits, attitudes and behaviours, as well as social demographic variables such as age, gender and education level. It was found that poor cognitive ability and poor attitudes towards transport safety were both positively correlated with unsafe behaviour; strong psychosocial beliefs were positively correlated with poor attitudes towards transport safety, but negatively correlated with unsafe behaviour. The study concludes with recommendations to improve traffic outcomes for the elderly.

Nighttime crashes involving older pedestrians pose a significant safety concern due to their age-related vulnerabilities such as reduced vision and slower reaction times. This study analyzes crash data from Texas for six years (2017-2022) using Association Rules Mining (ARM) to identify patterns and associations affecting crash severity for older pedestrians aged 65-74 years and those over 74 years under varying lighting conditions. The findings reveal that high-speed limits and complex road environments significantly increase the risk of fatal or severe injuries for both age groups, particularly under inadequate lighting. Additionally, demographic factors, adverse weather conditions, and specific road features further influence crash outcomes. These insights highlight the need for interventions, including lower speed limits, enhanced street lighting, and the implementation of advanced technologies such as modern pedestrian detection systems, sensor technology, pedestrian bags, accessible pedestrian signals, to improve the safety of older pedestrians. Policymakers should leverage these insights to formulate strategies that improve road safety for older pedestrians, addressing their unique vulnerabilities in various nighttime conditions.

In order to investigate whether retroreflective rings can enhance drivers' perception of spatial right-of-way in freeway tunnels, this paper explores a simulation test. The characteristics of spatial right-of-way in tunnels are elucidated, and a comparative test is conducted using commonly used delineators and raised pavement markers against retroreflective rings to enhance the perception of spatial right-of-way. The test employs the perception of lateral deviation and longitudinal distance as indicators to reflect the lateral and longitudinal right-of-way. Video scenarios, incorporating different facilities and spacing, are created using 3Ds Max software following the design standards of freeway tunnels. The indicators of Stimulation of Subjectively Equal Distance (SSED), lateral deviation, and perception reaction time (PRT) are chosen to assess the effects of different facilities on drivers under varying spacing conditions. Fifty-two participants, divided into two groups of novice drivers and experienced drivers, underwent perception testing in a simulated driving environment. The results indicate that drivers exhibit the highest overestimation of longitudinal distance and the longest PRT of lateral deviation in the absence of facilities. Installing retroreflective rings with a spacing of 50-200 m significantly mitigates the overestimation of longitudinal distance, while reducing the PRT of lateral deviation. On the other hand, setting up delineators and raised pavement markers with a spacing of 6-12 m significantly reduces the PRT of lateral deviation, while there is no significant enhancement to the perception of longitudinal distance. A spacing of 200 m for retroreflective rings and 10 m for delineators and raised pavement markers in the straight section is recommended as a safer and more economical setting scheme. The combination of these facilities can enhance drivers' safety perception of spatial right-of-way in freeway tunnels, facilitating rapid perception, correct judgment, and timely decision-making for the safe passage of vehicles.

Understanding the heterogeneity in autonomous vehicle (AV) crash patterns is crucial for enhancing the safety and public acceptance of autonomous transportation systems. In this paper, 584 AV collision reports from the California Department of Motor Vehicles (CA DMV) were first extracted and augmented by a highly automatic and fast variable extraction framework. Crash damage severities, classified as none, minor, moderate, and major, were set as the dependent variables. Factors including crash, road, temporal, vehicle, and environment characteristics were identified as potential determinants. To account for the heterogeneity inherent in crash data and identify key factors influencing the damage severity in AV crashes, a methodology integrating the latent class analysis and multinomial logit model was employed. Two heterogeneous clusters were determined based on the skewed distributions of vehicle status and driving mode. The model estimation results indicate a positive association between severe crash damage and some risk factors, such as head-on, intersection, multiple vehicles, dark with street lights, dark without street lights, and early morning. This study also reveals significant differences among the variables influencing the damage severity across two distinct subclasses. Moreover, partitioning the AV crash dataset into heterogeneous subsets facilitates the identification of critical factors that remain obscured when the dataset is analyzed as a whole, such as the evening indicator. This paper not only enhances our understanding of AV crash patterns but also paves the way for safer AV technology.

Anthropomorphic variation is an important factor in computational studies using Human Body Models (HBMs), particularly regarding how such differences can influence observed kinematics and loading. Currently, a gap exists between Anthropomorphic Test Devices (ATDs) and human body models (HBMs). By necessity, there are differences in constitutive behaviors at a material level, however segment mass distribution and anthropometry differences can make matched simulations of ATDs and HBMs difficult to interpret, which has real-world implications for current or future regulatory applications. In response to this gap, we present Global Human Body Model Consortium (GHBMC) 50th percentile male models (M50-OS) analogous to Hybrid III and THOR 50th percentile male models to serve as intermediaries between the HBMs and ATDs in computational biomechanics studies. Statistical human shape models were sized to each ATD, individually, using measurement metrics from ANSUR I. The M50-OS model surface and body shape surface were then converted to both polygon data (STL) and nodal coordinates. Landmark registration was performed, and the morphed nodes were imported back into the existing M50-OS models to generate two morphed versions analogous to the ATDs. The qualification suites were obtained from each respective ATD manual and run against the baseline and morphed models. The models were also simulated in a frontal NCAP crash pulses (56 km/hr) to assess variation in loading in a regulatory crash environment. 33 simulations were run between all models. The LSTC Hybrid III version 151,214 (open access software, ANSYS, Canonsburg, PA), Adult Shape Parametric Model (open access model, UMTRI, Ann Arbor, MI), GHBMC M50-OS (Elemance, LLC, Winston-Salem, NC), Thor v2.7 (University of Virginia School of Engineering and Applied Science/NHTSA) were used in this study. All simulations were run using LS-Dyna R. 9.3.1 (ANSYS) on an in-house computational cluster. The Male 50th Percentile THOR Analogous Occupant Simplified (M50 THAN-OS) has 364 × 10 elements and weighs 77.1 kg, the Hybrid III Analogous Occupant Simplified (M50 HAN-OS) has 370 × 10 elements and weighs 79.2 kg. The M50 HAN-OS and M50 THAN-OS response data exhibited a closer match to their respective models overall when compared to the baseline M50-OS human. In the vehicle test environment, the airbag and seatbelt peaks between M50 HAN-OS and M50 Hybrid III at nearly the same time. The results indicate that the ATD analogous models exhibit more similar response characteristics to their ATD counterparts after correcting for shape and mass distribution discrepancies. The ATD analogous models therefore represent a novel means to compare anticipated biomechanical loading on an occupant under similar testing conditions that use ATDs. Furthermore, the models represent a potential bridge towards testing modes of interest that ATDs are not well suited or are not validated for, such as low speed maneuvering, or oblique and far-side modes of crash.