Pub Date : 2026-01-20DOI: 10.1016/j.aap.2026.108415
Jie Wang , Lu Chen , Quankang Zhu , Shijian He , Amjad Pervez
Freeway ramps are recognized as high-risk segments of the road network due to their geometric complexity and dynamic traffic demands. This study investigates drivers’ mental workload in ramp areas by integrating psycho-physiological responses, specifically heart rate growth (HRG), with vehicle kinematic data, including speed and acceleration. Data were collected through real-world driving experiments from 32 experienced male drivers (aged 30–50 years) under both daytime and nighttime conditions. The findings revealed that HRG values were significantly higher at night, indicating increased cognitive stress in low-light conditions. In addition, the study identified a strong linear relationship between HRG and speed across all scenarios, indicating that increased speed is closely associated with higher mental workload. The relationship between HRG and acceleration followed a three-phase pattern, with sharp HRG changes at both low and high acceleration levels, and more stable responses within the mid-range. Based on these relationships, a classification framework was developed to categorize experienced male drivers’ mental workload into three workload categories (Class 1, Class 2, and Class 3) using joint thresholds of HRG, speed, and acceleration. These findings provide a data-driven basis for identifying cognitively demanding ramp segments and inform the design of adaptive speed guidance systems, real-time driver monitoring technologies, and ramp infrastructure improvements.
{"title":"Classifying experienced male drivers’ mental workload on freeway ramps based on heart rate and speed measurements: A real-vehicle experiment","authors":"Jie Wang , Lu Chen , Quankang Zhu , Shijian He , Amjad Pervez","doi":"10.1016/j.aap.2026.108415","DOIUrl":"10.1016/j.aap.2026.108415","url":null,"abstract":"<div><div>Freeway ramps are recognized as high-risk segments of the road network due to their geometric complexity and dynamic traffic demands. This study investigates drivers’ mental workload in ramp areas by integrating psycho-physiological responses, specifically heart rate growth (HRG), with vehicle kinematic data, including speed and acceleration. Data were collected through real-world driving experiments from 32 experienced male drivers (aged 30–50 years) under both daytime and nighttime conditions. The findings revealed that HRG values were significantly higher at night, indicating increased cognitive stress in low-light conditions. In addition, the study identified a strong linear relationship between HRG and speed across all scenarios, indicating that increased speed is closely associated with higher mental workload. The relationship between HRG and acceleration followed a three-phase pattern, with sharp HRG changes at both low and high acceleration levels, and more stable responses within the mid-range. Based on these relationships, a classification framework was developed to categorize experienced male drivers’ mental workload into three workload categories (Class 1, Class 2, and Class 3) using joint thresholds of HRG, speed, and acceleration. These findings provide a data-driven basis for identifying cognitively demanding ramp segments and inform the design of adaptive speed guidance systems, real-time driver monitoring technologies, and ramp infrastructure improvements.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108415"},"PeriodicalIF":6.2,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146017013","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}
Pub Date : 2026-01-20DOI: 10.1016/j.aap.2026.108411
Vincent Francoeur , Christine Saber , Steven Henderson , Charles Collin , Stephanie Yamin , Sylvain Gagnon
Peripheral motion contrast sensitivity decline is likely due to a progressive dysfunction of the magnocellular pathway in the aging brain. Previous research from our group had demonstrated that the Peripheral Motion Contrast Threshold 2-minute test version (PMCT-2) predicts older drivers’ hazardous behaviors in simulated driving environments. This study extends this work by examining correlations between PMCT-2 scores and on road driving outcomes at intersections coded from video recordings of fifty older drivers (65–89) navigating predefined urban routes in their own vehicles. We found significant correlations between PMCT-2 and scanning errors at non-signalized and stop-signalized intersections. We also found significant PMCT-2 correlation with driving compliance errors, notably incomplete stops, which was further supported by single-predictor regression using heteroskedasticity-robust estimation. Multiple linear regression analyses further showed that PMCT-2 remained the only significant predictor of stop-sign compliance errors after adjusting for age, gender and scanning error rates at stop signs. In contrast, its relationship with scanning errors was attenuated in linear models, reflecting the very low frequency of scanning errors observed on the road. These findings build on prior evidence that the PMCT-2 predicts older drivers’ performance outcomes and, for the first time, demonstrate its potential to predict actual on-road driving performance at intersections.
{"title":"Visual motion contrast thresholds in the periphery predict older drivers’ behavior at intersections","authors":"Vincent Francoeur , Christine Saber , Steven Henderson , Charles Collin , Stephanie Yamin , Sylvain Gagnon","doi":"10.1016/j.aap.2026.108411","DOIUrl":"10.1016/j.aap.2026.108411","url":null,"abstract":"<div><div>Peripheral motion contrast sensitivity decline is likely due to a progressive dysfunction of the magnocellular pathway in the aging brain. Previous research from our group had demonstrated that the Peripheral Motion Contrast Threshold 2-minute test version (PMCT-2) predicts older drivers’ hazardous behaviors in simulated driving environments. This study extends this work by examining correlations between PMCT-2 scores and on road driving outcomes at intersections coded from video recordings of fifty older drivers (65–89) navigating predefined urban routes in their own vehicles. We found significant correlations between PMCT-2 and scanning errors at non-signalized and stop-signalized intersections. We also found significant PMCT-2 correlation with driving compliance errors, notably incomplete stops, which was further supported by single-predictor regression using heteroskedasticity-robust estimation. Multiple linear regression analyses further showed that PMCT-2 remained the only significant predictor of stop-sign compliance errors after adjusting for age, gender and scanning error rates at stop signs. In contrast, its relationship with scanning errors was attenuated in linear models, reflecting the very low frequency of scanning errors observed on the road. These findings build on prior evidence that the PMCT-2 predicts older drivers’ performance outcomes and, for the first time, demonstrate its potential to predict actual on-road driving performance at intersections.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108411"},"PeriodicalIF":6.2,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146017068","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}
Pub Date : 2026-01-20DOI: 10.1016/j.aap.2026.108410
Yunxuan Li , Shihao Wang , Lishengsa Yue , Zhonghua Wei , Wenhui Zheng , Lin Zhang
High-fidelity vehicle trajectories are critical for proactive safety at intersections, where traditional reconstruction methods based on linear models or interpolation fail to capture complex nonlinear dynamics like sudden stops and sharp turns. While deep learning can model this complexity, its computational cost is prohibitive for real-time edge deployment. To address these challenges, this paper proposes an edge-computing-enhanced two-stage framework for high-fidelity trajectory reconstruction and dynamic risk assessment, specifically designed for Cooperative Vehicle-Infrastructure Systems (CVIS) at intersections. The first stage reconstructs accurate vehicle trajectories by applying physics-informed constraints derived from vehicle dynamics, combined with adaptive wavelet transforms and a hybrid thresholding strategy, enabling robust noise reduction from low-quality, multi-source sensor data. The second stage introduces a Vehicle Outline-based Conflict Algorithm (VOCA), which elevates traditional point-based conflict detection to outline-based spatial overlap analysis. By accurately modeling the real physical boundaries of vehicles, the proposed method significantly improves the sensitivity and timeliness of conflict detection, enabling more reliable proactive safety interventions in complex urban scenarios. Validated with real-world intersection data on an NVIDIA Jetson edge device, our method effectively suppresses high-frequency noise, reducing acceleration fluctuations by 98.66%. The outline-based VOCA proves vastly superior to traditional approaches, with center-point methods detecting only 22.53% of the conflicts identified by our algorithm. The entire framework achieves real-time performance, processing complex scenarios with delays under 100 ms per frame per vehicle. This work delivers an efficient solution for generating accurate, low-latency conflict warnings, advancing the practical application of CVIS for proactive safety management in urban environments.
{"title":"Proactive safety at CVIS-enabled intersections: a framework based on high-fidelity trajectory reconstruction and dynamic risk assessment","authors":"Yunxuan Li , Shihao Wang , Lishengsa Yue , Zhonghua Wei , Wenhui Zheng , Lin Zhang","doi":"10.1016/j.aap.2026.108410","DOIUrl":"10.1016/j.aap.2026.108410","url":null,"abstract":"<div><div>High-fidelity vehicle trajectories are critical for proactive safety at intersections, where traditional reconstruction methods based on linear models or interpolation fail to capture complex nonlinear dynamics like sudden stops and sharp turns. While deep learning can model this complexity, its computational cost is prohibitive for real-time edge deployment. To address these challenges, this paper proposes an edge-computing-enhanced two-stage framework for high-fidelity trajectory reconstruction and dynamic risk assessment, specifically designed for Cooperative Vehicle-Infrastructure Systems (CVIS) at intersections. The first stage reconstructs accurate vehicle trajectories by applying physics-informed constraints derived from vehicle dynamics, combined with adaptive wavelet transforms and a hybrid thresholding strategy, enabling robust noise reduction from low-quality, multi-source sensor data. The second stage introduces a Vehicle Outline-based Conflict Algorithm (VOCA), which elevates traditional point-based conflict detection to outline-based spatial overlap analysis. By accurately modeling the real physical boundaries of vehicles, the proposed method significantly improves the sensitivity and timeliness of conflict detection, enabling more reliable proactive safety interventions in complex urban scenarios. Validated with real-world intersection data on an NVIDIA Jetson edge device, our method effectively suppresses high-frequency noise, reducing acceleration fluctuations by 98.66%. The outline-based VOCA proves vastly superior to traditional approaches, with center-point methods detecting only 22.53% of the conflicts identified by our algorithm. The entire framework achieves real-time performance, processing complex scenarios with delays under 100 ms per frame per vehicle. This work delivers an efficient solution for generating accurate, low-latency conflict warnings, advancing the practical application of CVIS for proactive safety management in urban environments.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108410"},"PeriodicalIF":6.2,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146017096","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}
Pub Date : 2026-01-19DOI: 10.1016/j.aap.2026.108408
Monik Gupta, Nagendra R. Velaga
The design of the signal at the intersection considers the constant speed of the riders and the dilemma zone to be static. However, these assumptions may not hold true in complex environments with multiple users. This study explores the dynamic dilemma zone by incorporating the time to detect the signal by analyzing the drivers’ eye gaze movements and attention allocation patterns. The delay in detecting the amber phase of the signal can put drivers in a situation where they can neither safely cross the intersection nor stop before the stop line. The experiments were conducted in a virtual environment with 105 participants predominantly considering male riders. The image processing algorithms identified the first instance of riders noticing the amber phase. The parametric cure survival models were used to quantify the time to detect the signal as they incorporate the fact that some drivers may not look at the signal for the entire duration. This study further considered the complex decision-making of speeding and decelerating at the onset of amber phase at signalized intersections. The riders’ choices to vary the speed and safely or unsafely crossing the signal were quantified across psychological constraints. The results revealed that the odds of unsafe crossing at signal increased by 3.3, even in situations where riders were talking to pillion riders. The results indicated that riders under time pressure were more focused on the road, and their time to detect the signal was 0.72 s more than the base conditions.
{"title":"Dynamic dilemma zone at signalized intersection: attention allocation patterns using cure survival analysis for male riders","authors":"Monik Gupta, Nagendra R. Velaga","doi":"10.1016/j.aap.2026.108408","DOIUrl":"10.1016/j.aap.2026.108408","url":null,"abstract":"<div><div>The design of the signal at the intersection considers the constant speed of the riders and the dilemma zone to be static. However, these assumptions may not hold true in complex environments with multiple users. This study explores the dynamic dilemma zone by incorporating the time to detect the signal by analyzing the drivers’ eye gaze movements and attention allocation patterns. The delay in detecting the amber phase of the signal can put drivers in a situation where they can neither safely cross the intersection nor stop before the stop line. The experiments were conducted in a virtual environment with 105 participants predominantly considering male riders. The image processing algorithms identified the first instance of riders noticing the amber phase. The parametric cure survival models were used to quantify the time to detect the signal as they incorporate the fact that some drivers may not look at the signal for the entire duration. This study further considered the complex decision-making of speeding and decelerating at the onset of amber phase at signalized intersections. The riders’ choices to vary the speed and safely or unsafely crossing the signal were quantified across psychological constraints. The results revealed that the odds of unsafe crossing at signal increased by 3.3, even in situations where riders were talking to pillion riders. The results indicated that riders under time pressure were more focused on the road, and their time to detect the signal was 0.72 s more than the base conditions.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108408"},"PeriodicalIF":6.2,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146008747","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}
Pub Date : 2026-01-18DOI: 10.1016/j.aap.2025.108388
Yixin Shen , Hongfei Jia , Xin Ye , S.M. Lo , Biao He
Given the high efficiency and punctuality, transportation hub station are widely used by citizens, travelers daily. The large volume of passengers tends to cause overcrowding in transportation hub stations. Therefore, passenger movement efficiency has been a great concern for station designers, engineers, and facility managers. As the main facility connecting different floors in multilevel metro station, passengers’ movement on the vertical pedestrian transit facilities including stairs and escalators are critical for passengers’ safety and efficiency. To minimize passenger crowding and improve passenger movement efficiency, this study analysed the factors affecting passenger flow on vertical pedestrian transit facility and derived useful insights. By investigating the efficiency of passenger movement on the platform, influencing factors including the speed of escalator, passengers’ willingness to choose the stairs to move up floor levels, the layout and length of the mills barrier were explored. Furthermore, a safety-oriented evacuation layout was also detailed in the study. The study of the mills barrier revealed that a mills barrier placed between a staircase and an escalator promoted passenger efficiency. Moreover, a mills barrier length of 1 or 1.5 m is recommended. For the guidance strategy on the metro platform, the effect of passengers’ willingness to choose the stairs to move up on passenger efficiency was also investigated. Results indicated that passenger dwelling time decreased with an increasing proportion of passengers choosing the stairs. The suggested proportion of passengers choosing the stairs is 30%–40%, which effectively improve passenger efficiency. For the fire evacuation in transportation hub station, the removable facilities near the bottleneck point should be planned decently to be removed with the fastest speed, that will effectively speed up the evacuation process. The results are expected to be useful for designers, engineers, and facility managers.
{"title":"Safety-oriented facility design and operation management for transportation hub station","authors":"Yixin Shen , Hongfei Jia , Xin Ye , S.M. Lo , Biao He","doi":"10.1016/j.aap.2025.108388","DOIUrl":"10.1016/j.aap.2025.108388","url":null,"abstract":"<div><div>Given the high efficiency and punctuality, transportation hub station are widely used by citizens, travelers daily. The large volume of passengers tends to cause overcrowding in transportation hub stations. Therefore, passenger movement efficiency has been a great concern for station designers, engineers, and facility managers. As the main facility connecting different floors in multilevel metro station, passengers’ movement on the vertical pedestrian transit facilities including stairs and escalators are critical for passengers’ safety and efficiency. To minimize passenger crowding and improve passenger movement efficiency, this study analysed the factors affecting passenger flow on vertical pedestrian transit facility and derived useful insights. By investigating the efficiency of passenger movement on the platform, influencing factors including the speed of escalator, passengers’ willingness to choose the stairs to move up floor levels, the layout and length of the mills barrier were explored. Furthermore, a safety-oriented evacuation layout was also detailed in the study. The study of the mills barrier revealed that a mills barrier placed between a staircase and an escalator promoted passenger efficiency. Moreover, a mills barrier length of 1 or 1.5 m is recommended. For the guidance strategy on the metro platform, the effect of passengers’ willingness to choose the stairs to move up on passenger efficiency was also investigated. Results indicated that passenger dwelling time decreased with an increasing proportion of passengers choosing the stairs. The suggested proportion of passengers choosing the stairs is 30%–40%, which effectively improve passenger efficiency. For the fire evacuation in transportation hub station, the removable facilities near the bottleneck point should be planned decently to be removed with the fastest speed, that will effectively speed up the evacuation process. The results are expected to be useful for designers, engineers, and facility managers.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108388"},"PeriodicalIF":6.2,"publicationDate":"2026-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146002796","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}
Pub Date : 2026-01-17DOI: 10.1016/j.aap.2026.108406
Zeinab Bayati, Asad J. Khattak
Pedestrian safety remains a major concern, with fatalities rising despite infrastructure and safety improvements. To make meaningful progress, efforts should focus more intensely on reducing the most dangerous and fatal cases, given the growing importance of conventional and automated vehicle safety in shaping crash outcomes. This study introduces a composite unsupervised edge case detection framework that combines Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction with Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN). Each crash receives a composite score based on its cluster membership uncertainty and its distance from the core of typical crash patterns in the UMAP space. Based on these scores, crashes are classified into three interpretive layers: Core, Moderate Edge, and Strong Edge. Core cases represent common patterns, while Strong Edge cases reflect rare and complex situations. The framework is applied to 10,108 police-reported crashes from North Carolina coded with the Pedestrian and Bicycle Crash Analysis Tool (PBCAT), a relatively clean database of pedestrian crashes. Crash severity and contextual characteristics were compared across the three layers. Strong Edge crashes were substantially more severe, with 36.6% resulting in fatal injuries compared to 8.1% in the Core group. These high-risk cases often occurred in rural areas, under poor lighting conditions, in non-intersection locations, and involved behaviors such as unusual circumstances or crossing expressways. The findings show that the built environment and crash type influence pedestrian crash patterns. The edge case framework helps detect rare, high-risk crashes often missed by traditional methods, supporting targeted safety efforts.
{"title":"Beyond the norm: Identifying rare and high-risk pedestrian crash patterns using unsupervised learning","authors":"Zeinab Bayati, Asad J. Khattak","doi":"10.1016/j.aap.2026.108406","DOIUrl":"10.1016/j.aap.2026.108406","url":null,"abstract":"<div><div>Pedestrian safety remains a major concern, with fatalities rising despite infrastructure and safety improvements. To make meaningful progress, efforts should focus more intensely on reducing the most dangerous and fatal cases, given the growing importance of conventional and automated vehicle safety in shaping crash outcomes. This study introduces a composite unsupervised edge case detection framework that combines Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction with Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN). Each crash receives a composite score based on its cluster membership uncertainty and its distance from the core of typical crash patterns in the UMAP space. Based on these scores, crashes are classified into three interpretive layers: Core, Moderate Edge, and Strong Edge. Core cases represent common patterns, while Strong Edge cases reflect rare and complex situations. The framework is applied to 10,108 police-reported crashes from North Carolina coded with the Pedestrian and Bicycle Crash Analysis Tool (PBCAT), a relatively clean database of pedestrian crashes. Crash severity and contextual characteristics were compared across the three layers. Strong Edge crashes were substantially more severe, with 36.6% resulting in fatal injuries compared to 8.1% in the Core group. These high-risk cases often occurred in rural areas, under poor lighting conditions, in non-intersection locations, and involved behaviors such as unusual circumstances or crossing expressways. The findings show that the built environment and crash type influence pedestrian crash patterns. The edge case framework helps detect rare, high-risk crashes often missed by traditional methods, supporting targeted safety efforts.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108406"},"PeriodicalIF":6.2,"publicationDate":"2026-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974326","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}
Pub Date : 2026-01-16DOI: 10.1016/j.aap.2026.108409
Kaliprasana Muduli , Indrajit Ghosh , Satish V. Ukkusuri
Pedestrian safety remains a critical global concern, especially in countries like India, where unsignalized crossings with limited traffic control contribute to high pedestrian fatality rates. This study proposes a novel graph-based framework for analyzing pedestrian–vehicle interactions, advancing beyond traditional indicator-based, trajectory prediction, and pairwise modeling approaches. Indicator-based methods (e.g., PET, TTC) are retrospective and fail to capture evolving dynamics. Trajectory prediction models suffer from error accumulation, while pairwise models like LSTM or Transformer architectures are limited to two-agent interactions, restricting scalability and scene comprehension. In contrast, the proposed framework constructs holistic, scene-level multi-relational graphs, representing pedestrians and vehicles as interconnected nodes, with explicit modeling of pedestrian–pedestrian, vehicle–vehicle, and pedestrian–vehicle interactions. Unlike existing approaches that treat each pedestrian–vehicle pair in isolation, the proposed method represents all road users present in the scene as nodes in a dynamic spatio-temporal graph, enabling the model to learn not only direct pedestrian–vehicle relationships but also indirect influences mediated by surrounding pedestrians and vehicles. This design eliminates the need to pre-select interaction pairs, simplifying real-world deployment and improving scalability in dense traffic scenarios. Disentangled Multi-Scale Aggregation (DMSA) captures group behavior by focusing on contextually relevant agents, while a temporal CNN backbone models both short- and long-range dependencies efficiently. Empirical evaluations demonstrate the superior performance of the proposed model, which achieved a test accuracy of 90.6%, F1-score of 0.906, precision of 0.927, recall of 0.886, specificity of 0.928, and an AUC of 0.950, outperforming widely used baselines from the literature, such as GRU, LSTM, and Transformer-MLP, that have been applied in pedestrian interaction modeling tasks. Ablation studies confirmed the importance of Multi-Relational Adjacency Matrices (MRAM) and DMSA in improving accuracy and reducing false positives. By modeling scene-level dynamics, the framework enables context-aware prediction of critical events, supporting proactive conflict warning systems.
{"title":"A graph-based spatio-temporal framework for predicting safety-critical pedestrian–vehicle interactions at unsignalized crosswalks","authors":"Kaliprasana Muduli , Indrajit Ghosh , Satish V. Ukkusuri","doi":"10.1016/j.aap.2026.108409","DOIUrl":"10.1016/j.aap.2026.108409","url":null,"abstract":"<div><div>Pedestrian safety remains a critical global concern, especially in countries like India, where unsignalized crossings with limited traffic control contribute to high pedestrian fatality rates. This study proposes a novel graph-based framework for analyzing pedestrian–vehicle interactions, advancing beyond traditional indicator-based, trajectory prediction, and pairwise modeling approaches. Indicator-based methods (e.g., PET, TTC) are retrospective and fail to capture evolving dynamics. Trajectory prediction models suffer from error accumulation, while pairwise models like LSTM or Transformer architectures are limited to two-agent interactions, restricting scalability and scene comprehension. In contrast, the proposed framework constructs holistic, scene-level multi-relational graphs, representing pedestrians and vehicles as interconnected nodes, with explicit modeling of pedestrian–pedestrian, vehicle–vehicle, and pedestrian–vehicle interactions. Unlike existing approaches that treat each pedestrian–vehicle pair in isolation, the proposed method represents all road users present in the scene as nodes in a dynamic spatio-temporal graph, enabling the model to learn not only direct pedestrian–vehicle relationships but also indirect influences mediated by surrounding pedestrians and vehicles. This design eliminates the need to pre-select interaction pairs, simplifying real-world deployment and improving scalability in dense traffic scenarios. Disentangled Multi-Scale Aggregation (DMSA) captures group behavior by focusing on contextually relevant agents, while a temporal CNN backbone models both short- and long-range dependencies efficiently. Empirical evaluations demonstrate the superior performance of the proposed model, which achieved a test accuracy of 90.6%, F1-score of 0.906, precision of 0.927, recall of 0.886, specificity of 0.928, and an AUC of 0.950, outperforming widely used baselines from the literature, such as GRU, LSTM, and Transformer-MLP, that have been applied in pedestrian interaction modeling tasks. Ablation studies confirmed the importance of Multi-Relational Adjacency Matrices (MRAM) and DMSA in improving accuracy and reducing false positives. By modeling scene-level dynamics, the framework enables context-aware prediction of critical events, supporting proactive conflict warning systems.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108409"},"PeriodicalIF":6.2,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974327","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}
In transportation hubs, pedestrian flows form complex network structures, leading to serious congestion at peak hours. Understanding their dynamics is crucial to managing safe and efficient transportation. Although many experimental and theoretical studies have investigated pedestrian interactions at the microscopic level, computational models that account for pedestrians’ macroscopic origin and destination (OD) demands and mesoscopic route choices in large walking facilities are rare and lack empirical validation. In other words, pedestrians’ decision-making at strategic (macroscopic) and tactical (mesoscopic) levels, other than the operational (microscopic) level, has remained largely unexplored. Here, we propose an integrated Strategic–Tactical–Operational model for transportation hub (STO-Hub model), and validate it using 0.87 million pedestrian trajectories collected over three days by means of 11 LiDAR sensors at JR Shinjuku station in Japan. Based on an abstracted graph of the main concourse with directed links between different platform entrances and gates, we employ the gravity model at the strategic layer to estimate time-varying OD demand, a logit route-choice model at the tactical layer to capture route choice behavior, and an agent-based model to reproduce interactions with the surrounding environment and pedestrians. The STO-Hub model accurately reconstructs OD demand and route-choice behavior, achieving high agreement with directed flow counts, and the simulation delineates local congested areas evident in the sensing data. By estimating OD demand and route splits and by reproducing local interactions at any selected section, the STO-Hub model captures pedestrian dynamics across all three levels, including at congested locations. We further propose a STO-Hub framework that integrates sensing, the STO-Hub model, and management plans, providing a practical 10-min-resolution basis for OD-informed pedestrian guidance and control in transportation hubs. The study fills a gap in strategic modeling and management for large transportation hubs and supports congestion prevention, improved safety, and higher operational efficiency.
{"title":"Inferring the structure of pedestrian flows at a transportation hub","authors":"Xiaolu Jia , Claudio Feliciani , Hisashi Murakami , Sakurako Tanida , Liang Chen , Hao Yue , Daichi Yanagisawa , Katsuhiro Nishinari","doi":"10.1016/j.aap.2025.108391","DOIUrl":"10.1016/j.aap.2025.108391","url":null,"abstract":"<div><div>In transportation hubs, pedestrian flows form complex network structures, leading to serious congestion at peak hours. Understanding their dynamics is crucial to managing safe and efficient transportation. Although many experimental and theoretical studies have investigated pedestrian interactions at the microscopic level, computational models that account for pedestrians’ macroscopic origin and destination (OD) demands and mesoscopic route choices in large walking facilities are rare and lack empirical validation. In other words, pedestrians’ decision-making at strategic (macroscopic) and tactical (mesoscopic) levels, other than the operational (microscopic) level, has remained largely unexplored. Here, we propose an integrated Strategic–Tactical–Operational model for transportation hub (STO-Hub model), and validate it using 0.87 million pedestrian trajectories collected over three days by means of 11 LiDAR sensors at JR Shinjuku station in Japan. Based on an abstracted graph of the main concourse with directed links between different platform entrances and gates, we employ the gravity model at the strategic layer to estimate time-varying OD demand, a logit route-choice model at the tactical layer to capture route choice behavior, and an agent-based model to reproduce interactions with the surrounding environment and pedestrians. The STO-Hub model accurately reconstructs OD demand and route-choice behavior, achieving high agreement with directed flow counts, and the simulation delineates local congested areas evident in the sensing data. By estimating OD demand and route splits and by reproducing local interactions at any selected section, the STO-Hub model captures pedestrian dynamics across all three levels, including at congested locations. We further propose a STO-Hub framework that integrates sensing, the STO-Hub model, and management plans, providing a practical 10-min-resolution basis for OD-informed pedestrian guidance and control in transportation hubs. The study fills a gap in strategic modeling and management for large transportation hubs and supports congestion prevention, improved safety, and higher operational efficiency.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108391"},"PeriodicalIF":6.2,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974324","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}
Pub Date : 2026-01-16DOI: 10.1016/j.aap.2026.108407
Jiyao Wang , Wenbo Li , Zhenyu Wang , Suzan Ayas , Birsen Donmez , Dengbo He , Kaishun Wu
Driver drowsiness is one of the leading causes of crashes, injuries, and fatalities on the road. Traditional drowsiness detection models relied on manually extracted physiological features processed through machine learning algorithms. However, these methods lacked flexibility and robustness across diverse real-world conditions. Although recent advances in deep learning have improved detection accuracy through automated feature extraction based on larger learnable parameter space, the generalization of existing models is still limited due to domain shifts. In this study, we proposed DrowsyDG-Phys, a novel domain generalization (DG) framework for driver drowsiness detection using three physiological signals (i.e., electrocardiogram, electrodermal activity, and respiration signals) that can be measured by in-vehicle or wearable sensors. Our approach introduced a backbone network for explicit time and frequency domain feature learning. In addition, our approach integrated three novel loss functions: a prior knowledge-based contrastive regularization for robustness, a feature centralization loss to promote generalization in heterogeneities, and a novel loss function to align drowsiness assessment criteria. Finally, we established a multi-source DG benchmark and evaluated our model on three existing datasets and a self-collected dataset involving 60 participants in a simulated SAE Level-3 driving scenario. Our proposed DrowsyDG-Phys achieves 78.5% accuracy on the DG protocol, as well as 88.4% accuracy on the cross-subject protocol. Experimental results demonstrated that DrowsyDG-Phys outperformed baseline methods, and improved generalization and robustness of physiological signal-based drowsiness monitoring.
{"title":"DrowsyDG-Phys: Generalizable driver drowsiness estimation in conditional automated vehicles using physiological signals","authors":"Jiyao Wang , Wenbo Li , Zhenyu Wang , Suzan Ayas , Birsen Donmez , Dengbo He , Kaishun Wu","doi":"10.1016/j.aap.2026.108407","DOIUrl":"10.1016/j.aap.2026.108407","url":null,"abstract":"<div><div>Driver drowsiness is one of the leading causes of crashes, injuries, and fatalities on the road. Traditional drowsiness detection models relied on manually extracted physiological features processed through machine learning algorithms. However, these methods lacked flexibility and robustness across diverse real-world conditions. Although recent advances in deep learning have improved detection accuracy through automated feature extraction based on larger learnable parameter space, the generalization of existing models is still limited due to domain shifts. In this study, we proposed <strong>DrowsyDG-Phys</strong>, a novel domain generalization (DG) framework for driver drowsiness detection using three physiological signals (i.e., electrocardiogram, electrodermal activity, and respiration signals) that can be measured by in-vehicle or wearable sensors. Our approach introduced a backbone network for explicit time and frequency domain feature learning. In addition, our approach integrated three novel loss functions: a prior knowledge-based contrastive regularization for robustness, a feature centralization loss to promote generalization in heterogeneities, and a novel loss function to align drowsiness assessment criteria. Finally, we established a multi-source DG benchmark and evaluated our model on three existing datasets and a self-collected dataset involving 60 participants in a simulated SAE Level-3 driving scenario. Our proposed DrowsyDG-Phys achieves 78.5% accuracy on the DG protocol, as well as 88.4% accuracy on the cross-subject protocol. Experimental results demonstrated that DrowsyDG-Phys outperformed baseline methods, and improved generalization and robustness of physiological signal-based drowsiness monitoring.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108407"},"PeriodicalIF":6.2,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974338","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}
Pub Date : 2026-01-16DOI: 10.1016/j.aap.2026.108402
Rui Li, Yiru Liu, Jian Sun, Ye Tian
With the large-scale deployment of autonomous vehicles (AVs), AV-human-driven vehicle (HV) interactions are increasingly common. AVs face a social dilemma: competitive behavior raises ethical and public acceptance concerns, whereas cooperative behavior can invite exploitation and degrade efficiency. We use Evolutionary Game Theory (EGT) to model long-run adaptation between AVs and HVs and quantify agent sociality via a data-calibrated Social Value Orientation (SVO) metric. After calibrating HV social preferences from unprotected left-turn trajectories, we incorporate HV heterogeneity into a two-population EGT with cooperative and competitive types. SVO-informed rewards are used to construct payoff matrices for replicator analyses to identify evolutionarily stable strategies (ESS). Experiments show that AV policies with moderate egoism mitigate the social dilemma and tend to achieve population-level dominance in both roles (left-turning and straight-going), whereas overly cooperative policies are evolutionarily unstable. Moreover, AVs benefit from opponent-aware, dynamically adjustable sociality to accommodate diverse HV preferences. To test the theory, we run agent-based imitation simulations. Sensitivity analyses indicate that AV advantages are hard to observe at low market penetration but become pronounced as penetration approaches about 50%, after which convergence accelerates. Overall, the framework clarifies when and why AV sociality preferences succeed over time, offering actionable guidance for designing adaptive, socially compatible AV decision policies in mixed traffic.
{"title":"Cooperative or competitive? Resolving social dilemmas in autonomous vehicles through evolutionary game theory","authors":"Rui Li, Yiru Liu, Jian Sun, Ye Tian","doi":"10.1016/j.aap.2026.108402","DOIUrl":"10.1016/j.aap.2026.108402","url":null,"abstract":"<div><div>With the large-scale deployment of autonomous vehicles (AVs), AV-human-driven vehicle (HV) interactions are increasingly common. AVs face a social dilemma: competitive behavior raises ethical and public acceptance concerns, whereas cooperative behavior can invite exploitation and degrade efficiency. We use Evolutionary Game Theory (EGT) to model long-run adaptation between AVs and HVs and quantify agent sociality via a data-calibrated Social Value Orientation (SVO) metric. After calibrating HV social preferences from unprotected left-turn trajectories, we incorporate HV heterogeneity into a two-population EGT with cooperative and competitive types. SVO-informed rewards are used to construct payoff matrices for replicator analyses to identify evolutionarily stable strategies (ESS). Experiments show that AV policies with moderate egoism mitigate the social dilemma and tend to achieve population-level dominance in both roles (left-turning and straight-going), whereas overly cooperative policies are evolutionarily unstable. Moreover, AVs benefit from opponent-aware, dynamically adjustable sociality to accommodate diverse HV preferences. To test the theory, we run agent-based imitation simulations. Sensitivity analyses indicate that AV advantages are hard to observe at low market penetration but become pronounced as penetration approaches about 50%, after which convergence accelerates. Overall, the framework clarifies when and why AV sociality preferences succeed over time, offering actionable guidance for designing adaptive, socially compatible AV decision policies in mixed traffic.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"228 ","pages":"Article 108402"},"PeriodicalIF":6.2,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974328","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号