Accident; analysis and prevention最新文献

{"title":"Out-of-sample prediction and interpretation for random parameter generalized linear models","authors":"Jonathan S. Wood ,&nbsp;Vikash Gayah","doi":"10.1016/j.aap.2025.108147","DOIUrl":"10.1016/j.aap.2025.108147","url":null,"abstract":"<div><div>Incorporating random parameters (RPs) into generalized linearized models (GLMs) – such as the negative binomial (NB) regression model used to predict crash frequencies – has been shown to improve model fit and better address issues such as unobserved heterogeneity. However, applying models with RPs to make predictions for observations outside the sample used to estimate the model is not straightforward. Recent studies have proposed various methods to incorporate RPs in out-of-sample predictions, but these tend to provide biased estimates or are computationally intensive to apply. This paper applies fundamental statistical theory to leverage properties of the underlying RP distributions incorporated into GLMs to provide more direct and accurate predictions, as well as directly estimate prediction variance for out-of-sample observations. Methods are provided for several common RP distributions – including the normal/Gaussian, lognormal, triangular, uniform, and gamma distributions – combined within log-link GLM framework. Additionally, closed-form equations for elasticities and marginal effects for the random parameters are provided. The proposed methods are tested using crash frequency prediction models developed using data from the Highway Safety Information System (HSIS). The results suggest that the proposed exact method provides more accurate predictions than the computational-intensive simulation-based approximation approaches while also being simple to apply. The method is suitable for the widespread use of RPs in research and in practical applications of GLMs.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108147"},"PeriodicalIF":5.7,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144588172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Micro-level hotspot identification at intersections using traffic conflict analysis","authors":"Nuri Park ,&nbsp;Juneyoung Park ,&nbsp;Yang-Jun Joo ,&nbsp;Mohamed Abdel-Aty","doi":"10.1016/j.aap.2025.108167","DOIUrl":"10.1016/j.aap.2025.108167","url":null,"abstract":"<div><div>Traditional approaches to identifying traffic crash hotspots have mainly focused on determining dangerous intersections within road networks, overlooking variations in crash risk within intersections. The micro-level crash hotspot analysis addresses this issue by identifying specific high-risk areas with precision. This study aims to identify micro-level hotspots within three signalized intersections using traffic conflict measures derived from drone video. An algorithm calculates conflicts based on various vehicle sizes and conflict angles. The traffic conflict measures in this study include time-to-collision (TTC), the time to a potential collision assuming constant speed; modified time-to-collision (MTTC), which detects conflicts by assuming constant acceleration; and post-encroachment time (PET), the time gap between two vehicles passing the same point. To select the most appropriate conflict measures and determine optimal thresholds at each intersection, we develop crash frequency models using generalized linear modeling (GLM). These selected conflict measures and thresholds are subsequently used to detect micro-level hotspot sections through kernel density. The results demonstrate that the TTC and PET are strongly related to micro-level crash frequencies, with different patterns emerging depending on crash angle and intersection location. Specifically, TTC-based conflicts are highly correlated with rear-end crashes occurring before the stop line, while PET-based conflicts are closely associated with crashes within the intersection, particularly with left-turning movements. This study contributes to intersection safety by identifying traffic conflict measures for micro-level hotspots and offering detailed safety interventions. These interventions include pavement marking enhancements, stop-line location adjustment, extended left-turn bays, or separated bike lanes, which are based on the specific conflict patterns observed in the study.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108167"},"PeriodicalIF":5.7,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144588063","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A spatial generalized extreme value framework for traffic conflict using max-stable process approach","authors":"Quansheng Yue ,&nbsp;Yanyong Guo ,&nbsp;Tarek Sayed ,&nbsp;Pan Liu ,&nbsp;Hao Lyu ,&nbsp;Wentao Fan","doi":"10.1016/j.aap.2025.108164","DOIUrl":"10.1016/j.aap.2025.108164","url":null,"abstract":"<div><div>Extreme value theory (EVT) models are widely used to estimate crash risk from traffic conflicts for proactive traffic safety management. However, existing EVT models assume that the crash risks are evenly distributed across the entire study area, ignoring the spatial effect across different zones within the area. This study proposes a spatial EVT modeling framework using max-stable process (MSP) approach for traffic conflicts while accounting for spatial dependence. Traffic conflict data from vehicle trajectories on U.S.101, sourced from the NGSIM dataset, were utilized with time to collision (TTC) as the conflict indicator. Three types of MSP models are used to capture spatial dependence: Schlather, Brown-Resnick, and Smith, each with corresponding correlation functions. Various correlation functions and link functions for each MSP model were proposed. The pairwise composite likelihood estimation approach is utilized to estimate the MSP models’ parameters, and the extremal coefficient indicator is employed to describe the spatial dependence across different zones. Crash risk is estimated for each zone within the study area. Model results show significant spatial correlation in extreme traffic conflicts. Moreover, spatial dependence in these extreme conflicts diminishes with distance, showing stronger correlations at shorter distances. M1 achieved the best goodness-of-fit among the MSP models, indicates that the integration of spatial covariates in the threshold and scale parameters effectively explains a significant amount of variation in the observed data. In particular, the Schlather model with a powered exponential correlation function performs better than the Smith and Brown-Resnick models. The crash risk analysis result shows that inner (fast) lanes have lower crash risk than outer lanes, and crash risk is higher on the entrance ramp than the exit ramp. The crash risk estimated from the spatial EVT model is consistent with the TTC heatmap, particularly in high conflict zones, demonstrating the reliability and validity of the spatial EVT modeling approach for traffic safety analysis.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108164"},"PeriodicalIF":5.7,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144596373","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of information-assisted modalities and levels on driver takeover performance in conditional automated driving","authors":"Zhenhua Yu ,&nbsp;Linjing Zhai ,&nbsp;Kang Jiang ,&nbsp;Shuangyu Yu ,&nbsp;Bingzhan Zhang ,&nbsp;Qingqing Deng ,&nbsp;Zhipeng Huang","doi":"10.1016/j.aap.2025.108158","DOIUrl":"10.1016/j.aap.2025.108158","url":null,"abstract":"<div><div>With advances in autonomous driving technology, conditional automated driving (CAD) is gradually entering the consumer market. Before full automation is achieved, however, drivers are still required to take control of the vehicle in complex scenarios. To increase safety and comfort in CAD systems, drivers must be able to respond promptly to takeover requests (TORs). While previous studies have focused on warning signals, it remains unclear how to provide supportive information for complex operations during TORs to optimize post-takeover behavior. This study designed and evaluated a multimodal in-vehicle information assistance system to support safer driver takeovers by providing real-time road information, such as hazard warnings and lane availability. The study introduced two independent variables, three information assistance modalities (visual (V), auditory (A), and visual + auditory (V + A)) and two levels of assistance information (hazard information (H) and hazard information + operational suggestions (HO)). A driving simulator experiment with 56 participants assessed the effects of these conditions on driving performance, gaze behavior, and subjective evaluation. Results demonstrated that the V + A modality combined with HO information achieved the best performance in takeover speed and lane-change safety. Compared to the baseline, takeover time was reduced by 0.46 s, and time-to-collision (TTC) improved by 3.59 s, reducing collision risk. The visual (V) modality enhanced vehicle stability, lowering maximum resultant acceleration by 0.73 m/s² and reducing abrupt maneuvers. The HO condition improved decision-making quality, increasing overtaking success rates by 6.25 % without adding cognitive load. Additionally, the V + A and HO combination optimized attention allocation. Multisensory integration (V + A) enhanced environmental awareness and decision speed by minimizing information omission through simultaneous visual and auditory cues. These findings provide key insights into designing information assistance for future automated driving, emphasizing the importance of optimizing information delivery in emergencies to improve safety and user experience.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108158"},"PeriodicalIF":5.7,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144569825","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Safety performance of passing zone segments on two-lane rural highways in Pennsylvania: Comparing crash modification factors from causal inference and unobserved heterogeneity models","authors":"Prakash Poudel,&nbsp;Eric T. Donnell,&nbsp;Vikash V. Gayah","doi":"10.1016/j.aap.2025.108165","DOIUrl":"10.1016/j.aap.2025.108165","url":null,"abstract":"<div><div>Passing zones on two-lane rural highways are marked based on minimum passing sight distance criteria associated with the 85th-percentile speed or posted speed limit along the roadway segment. Although passing-related crashes account for a relatively small proportion of total reported crashes on two-lane rural highways, past research suggests that they tend to result in more severe injuries than non-passing-related crashes. However, the safety performance of roadway segments with passing zones has not been quantified or compared to segments with no passing zones. The purpose of this paper is to use data from Pennsylvania to compare the safety performance of two-lane rural highways with and without the presence of passing zone markings. Total crashes, fatal plus injury crashes, and target crashes are used to estimate crash modification factors (CMFs) for the presence of passing zones. A second objective of the paper is to compare the CMFs developed using two different methodological approaches: the propensity scores-potential outcomes causal inference framework and unobserved heterogeneity (random parameters) models. The results indicate that the CMFs developed using these two approaches are similar, although the CMFs from the unobserved heterogeneity models tended to estimate slightly fewer expected crashes in passing zone segments than the causal inference method. When compared to road segments without passing zones, those with passing zones experienced fewer total crashes, fatal and injury crashes, and head-on plus sideswipe crashes by 11.2 %, 12.2 %, and 10.6 %, respectively, based on the causal inference method.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108165"},"PeriodicalIF":5.7,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144569828","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A dynamic and stochastic perspective on time headway in traffic oscillations and its implications for traffic safety","authors":"Ying Luo ,&nbsp;Weijie Yu ,&nbsp;Jiaxin Lu ,&nbsp;Yanyan Chen ,&nbsp;Dong Ngoduy","doi":"10.1016/j.aap.2025.108146","DOIUrl":"10.1016/j.aap.2025.108146","url":null,"abstract":"<div><div>Traffic oscillations refer to the alternating patterns of vehicle deceleration and acceleration in congested conditions, which usually create significant safety concerns on freeways. Thus, it is imperative to understand the mechanisms of traffic oscillations and their underlying safety implications. This paper presents a novel approach to exploring the combined effects of dynamic time headway (DTH) and stochasticity on traffic oscillations during car-following. Using high-precision trajectory data, we demonstrate a strong correlation between DTH and stochasticity strength with the power functions of speed. We then extend the car-following model framework that considers both the dynamic characteristics and stochasticity of time headway to investigate the mechanisms of traffic oscillation. The model calibration and validation results demonstrate that our extended model outperforms the original model in terms of trajectory fitting accuracy, successfully replicating the asymmetric driving behavior and the concave growth pattern of speed standard deviation. Building upon this novel perspective, linear stability and safety evaluation are systematically conducted to understand the comprehensive influence of DTH and stochasticity. Our theoretical and numerical experiments show that DTH significantly increases the range of string instability in traffic flow, particularly at low-speed regimes. The influence of the stochasticity on the marginal stability of traffic flow shows a pattern of increasing followed by decreasing tendencies. Also, the combined effect of drivers’ DTH characteristics and stochasticity could expand the rear-end collision risks at low-speed regimes, showing a backward diffusion effect. Our findings further establish the interconnection of traffic oscillations with traffic stability and safety concerns.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108146"},"PeriodicalIF":5.7,"publicationDate":"2025-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144569826","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trajectory-based indicators to determine the local character of intersection conflicts: A micro-spatial analysis","authors":"Andreas Hula ,&nbsp;Apostolos Ziakopoulos ,&nbsp;Ángel Losada ,&nbsp;Andrea Schaub ,&nbsp;Peter Saleh ,&nbsp;George Yannis","doi":"10.1016/j.aap.2025.108155","DOIUrl":"10.1016/j.aap.2025.108155","url":null,"abstract":"<div><div>Real world behaviour data is the most reliable reference to assess road safety in a specific road infrastructure context. However, its collection and implementation for road safety research in a rapid and portable manner is still challenging, facing data protection issues and the complexities to set up constant tracking mechanisms with their own power supply. To tackle these limitations, the Mobility Observation Box (MOB) provides a flexible data collection, to be used in subsequent video analysis. Object detection and tracking allow for the derivation of movement trajectories, which in turn allow to derive quantitative indicators of road safety relevant behaviour, namely well-established Surrogate Safety Measures (SSMs), such as Post-Encroachment-Time (PET), Time-to-Closest-Approach (TCA) and Time-to-Collision-(TTC) alongside a number of indicators like maximum speed of two interaction partners, the angle they approach each other in and the minimum distance they had at one point in their interaction. To facilitate potential MOB uses, this study leverages over 51 h of naturalistic video data at a busy Vienna intersection to advance road safety research by (i) employing random parameters binary modelling of the likelihood of critical conflict occurrence and (ii) Gaussian generalized additive spatial modelling to identify key factors influencing the absolute values of conflict angles on critical conflicts only. Within the examined intersection, specific speed and acceleration effects were determined, together with the respective heterogeneity-in-means, as well as significant categorical effects of different road user types. All road user types were ultimately less likely to be involved in safety–critical conflicts compared to cars in both leading (firstly detected) and following (secondly detected) roles, with the exception of cyclists in the leading role. Within the micro-spatial analysis, the kinematic parameters of the second road user only (speed, max acceleration and max deceleration), the duration of the interaction as well as intersection-specific local effects related to the position of the leading road user were all found to influence the transformed absolute value of the angles of critical conflicts.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108155"},"PeriodicalIF":5.7,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144548571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic evaluation of emergency lane occupation based on an improved driving risk field model","authors":"Chuzhao Li ,&nbsp;Yangming Zhou ,&nbsp;Zhiyu Tao ,&nbsp;Guofa Li ,&nbsp;Zhou Wu","doi":"10.1016/j.aap.2025.108143","DOIUrl":"10.1016/j.aap.2025.108143","url":null,"abstract":"<div><div>In China, the frequent occupation of emergency lanes on highways significantly affects traffic efficiency and safety. This study collected data from six highway segments and extracted lane-changing (LC) behavior. A logistic regression model was employed to identify key factors influencing emergency lane changes (ELC). An improved driving risk field model was developed to assess risks during the LC process, integrating the heading angle of the obstacle vehicles, speeds of both the main and obstacle vehicles, as well as the filtering effect of lane lines. Compared to three existing risk field models, our model provides a more accurate assessment of collision risks during lane changes. This study revealed that the primary reasons for ELC are driven by the pursuit of speed and driving space, influenced by the speed and relative distance of surrounding vehicles, and the intention to overtake. Moreover, ELC is characterized by reduced lateral speed and acceleration, as well as a lower minimum time-to-collision (TTC), while exhibiting higher acceptable risks. Risk variation trends during lane changes from emergency to normal lanes include conservative, general, and aggressive types, while changes from normal to emergency lanes feature only conservative and general types. Additionally, ELC significantly negatively impacts following vehicles in the target lane after LC, leading to greater lateral displacement, more significant longitudinal deceleration, and heightened risk levels.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108143"},"PeriodicalIF":5.7,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144535839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A secure and privacy-preserving authentication framework for Connected and Autonomous Vehicles based on DRG-PBFT and zero-knowledge proof","authors":"Anjum Mohd Aslam,&nbsp;Aditya Bhardwaj,&nbsp;Rajat Chaudhary","doi":"10.1016/j.aap.2025.108145","DOIUrl":"10.1016/j.aap.2025.108145","url":null,"abstract":"<div><div>Connected and Autonomous Vehicles (CAVs) are pivotal to advancing Intelligent Transportation Systems (ITS) but introduce significant security and privacy challenges, particularly in dynamic environments requiring real-time data exchange. The existing security measures and consensus mechanisms, such as Practical Byzantine Fault Tolerance (PBFT), are susceptible to various attacks, including identity forgery, unauthorized access, and compromised safety testing, and suffer from scalability and latency issues. This study addresses these challenges by proposing a Dynamic Reputation Grouping-based PBFT (DRG-PBFT) approach, integrated with Simulation Extractable Zero-Knowledge Succinct Non-Interactive Arguments of Knowledge (SE-ZK-SNARKS). The proposed framework leverages reputation-based dynamic grouping to enhance consensus efficiency and reduce communication overhead. SE-ZK-SNARKS provide anonymity and privacy-preserving identity authentication, enabling CAVs to prove their legitimacy without revealing sensitive information. The proposed framework has been validated through extensive simulations using the NS-3 network simulator integrated with blockchain. The simulation results demonstrate that our proposed approach outperforms existing methods, achieving reduced consensus latency, communication overhead, authentication time and improved throughput. Overall, the findings and methodologies presented in this study address critical challenges in securing CAV communications while maintaining scalability and efficiency and can serve as a valuable reference for researchers and practitioners aiming to improve the safety and reliability of CAVs in real-time environments.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108145"},"PeriodicalIF":5.7,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144535838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Human-Centric explanations for users in automated Vehicles: A systematic review","authors":"Zishuo Zhu ,&nbsp;Xiaomeng Li ,&nbsp;Patricia Delhomme ,&nbsp;Ronald Schroeter ,&nbsp;Sebastien Glaser ,&nbsp;Andry Rakotonirainy","doi":"10.1016/j.aap.2025.108152","DOIUrl":"10.1016/j.aap.2025.108152","url":null,"abstract":"<div><h3>Background</h3><div>The decision-making processes of automated vehicles (AVs) can confuse users and reduce trust, highlighting the need for clear and human-centric explanations. Such explanations can help users understand AV actions, facilitate smooth control transitions and enhance transparency, acceptance, and trust. Critically, such explanations could improve situational awareness and support timely, appropriate human responses, thereby reducing the risk of misuse, unexpected automated decisions, and delayed reactions in safety–critical scenarios. However, current literature offers limited insight into how different types of explanations impact drivers in diverse scenarios and the methods for evaluating their quality. This paper systematically reviews what, when and how to provide human-centric explanations in AV contexts.</div></div><div><h3>Methods</h3><div>The systematic review followed PRISMA guidelines, and covered five databases—Scopus, Web of Science, IEEE Xplore, TRID, and Semantic Scholar—from 2000 to April 2024. Out of 266 identified articles, 59 met the inclusion criteria.</div></div><div><h3>Results</h3><div>Providing a detailed content explanation following AV’s driving actions in real time does not always increase user trust and acceptance. Explanations that clarify the reasoning behind actions are more effective than those merely describing actions. Providing explanations before action is recommended, though the optimal timing remains uncertain. Multimodal explanations (visual and audio) are most effective when each mode conveys unique information; otherwise, visual-only explanations are preferred. The narrative perspective (first-person vs. third-person) also impacts user trust differently across scenarios.</div></div><div><h3>Conclusions</h3><div>The review underscores the importance of tailoring human-centric explanations to specific driving contexts. Future research should address explanation length, timing, and modality coordination and focus on real-world studies to enhance generalisability. These insights are vital for advancing the research of human-centric explanations in AV systems and fostering safer, more trustworthy human-vehicle interactions, ultimately reducing the risk of inappropriate reactions, delayed responses, or user error in traffic settings.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"220 ","pages":"Article 108152"},"PeriodicalIF":5.7,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}