Transportation Research Part C-Emerging Technologies最新文献

{"title":"Modelling riders’ intervention behavior during high-level autonomous driving under extreme conditions","authors":"Zheng Xu,&nbsp;Nan Zheng,&nbsp;Yihai Fang,&nbsp;Hai.L Vu","doi":"10.1016/j.trc.2025.105367","DOIUrl":"10.1016/j.trc.2025.105367","url":null,"abstract":"<div><div>The development of autonomous driving systems (ADS) has primarily focused on technical advancements to prevent accidents and enhance overall transport performance. While significant strides have been made in improving the performance of autonomous vehicles (AVs), there remains a substantial disconnect between the intelligence of AVs and user acceptance. This study aims to illuminate the differences in decision-making between the ADS and riders in AVs during extreme crash scenarios, identifying the specific factors that prompt rider interventions. We recreated three typical fatal road accidents from Australian roads, within a high-fidelity virtual reality (VR) environment. In these simulations, vehicles involved in the original crashes were replaced with fine-tuned level 4 AVs to evaluate whether the accidents would occur in a similar manner. We engaged human participants from diverse demographic backgrounds in a human-in-the-loop analysis, immersing them in these scenarios by sitting in the simulated AVs to gather insights into their perceptions and reactions. Our study investigated the factors influencing riders’ intervention behaviors and highlighted the decision-making disparities between ADS and human riders. We quantified the nature of these interventions during autonomous driving by defining turning and accelerating indices. The results revealed a strong correlation between interventions and vehicle movement, with intervention probabilities exceeding 80% when the AV’s acceleration index surpasses 0.7. Most importantly, our findings distinguished between unnecessary and necessary interventions during rider interactions with ADS under extreme conditions. We showed that necessary interventions can help refine ADS maneuvers at intersections by tempering aggressive responses, offering valuable guidance for system development. These insights not only inform current ADS enhancement strategies but pave the way for future research aimed at reducing unnecessary interventions while recognizing the value of necessary ones, ultimately supporting the broader adoption of high-level ADS.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"181 ","pages":"Article 105367"},"PeriodicalIF":7.6,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145247942","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":"Detecting shifts in urban rail passenger behavior during emergencies: A data-driven comparative approach","authors":"Qiuchi Xue ,&nbsp;Xin Yang ,&nbsp;Xiqun (Michael) Chen ,&nbsp;Jianjun Wu ,&nbsp;Ziyou Gao","doi":"10.1016/j.trc.2025.105377","DOIUrl":"10.1016/j.trc.2025.105377","url":null,"abstract":"<div><div>During urban rail transit (URT) emergencies, passengers often modify their travel behavior in response to perceived risks and disruptions. Detecting and understanding these behavioral shifts is essential for evaluating the impacts of such events and designing effective contingency strategies. This paper proposes a data-driven approach to detect individual behavior changes by comparing observed behavior during emergencies with predicted normal behavior. To establish reliable references, we design a continuous rolling prediction pipeline that dynamically forecasts routine travel patterns, serving as counterfactual baselines for comparison. We validate our approach through a case study using data from Beijing’s URT system. Results demonstrate that the predictive module improves the accuracy of normal individual travel behavior predictions, and our data-driven approach effectively detects behavioral changes during emergencies. These findings offer valuable insights for understanding individual behavior variability, improving emergency planning, and forecasting passenger flow under crisis conditions.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"181 ","pages":"Article 105377"},"PeriodicalIF":7.6,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145271508","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":"Decision making for autonomous vehicles: A mixed curriculum reinforcement learning approach and a novel safety intervention method","authors":"Hongqing Chu ,&nbsp;Heng Wang ,&nbsp;Yifan Cheng ,&nbsp;Aoyong Li ,&nbsp;Wei Tian ,&nbsp;Bingzhao Gao ,&nbsp;Hong Chen","doi":"10.1016/j.trc.2025.105369","DOIUrl":"10.1016/j.trc.2025.105369","url":null,"abstract":"<div><div>Reinforcement learning is considered one of the most promising approaches for decision-making in autonomous vehicles within interactive scenarios. However, its implementation faces challenges of insufficient safety and limited learning efficiency due to the stochastic nature of exploration and the complexity of the exploration space. In this paper, a mixed curriculum learning (MCL) approach, incorporating an intervention method called discrepancy-directed Bernoulli intervention (DDBI), is proposed to address these challenges in reinforcement learning. Firstly, the algorithm divides the training process into a safety phase and a performance phase. The agent focuses on accomplishing the safety task first, which is followed by the performance task. Secondly, DDBI introduces an additional safety agent to intervene in hazardous situations using a novel probability-based method, thereby enhancing the safety of the training process while preserving the exploratory nature of reinforcement learning. Finally, the proposed approach is evaluated in a lane change scenario with random traffic flow. Comprehensive comparative experiments with other algorithms demonstrate that the proposed approach outperforms in both safety and learning efficiency.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"181 ","pages":"Article 105369"},"PeriodicalIF":7.6,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145271509","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":"Tactical demand and capacity balancing using incremental search in spatio-temporal graphs with flight uncertainty","authors":"Yutong Chen ,&nbsp;Ramon Dalmau ,&nbsp;Sameer Alam","doi":"10.1016/j.trc.2025.105382","DOIUrl":"10.1016/j.trc.2025.105382","url":null,"abstract":"<div><div>Demand and Capacity Balancing (DCB) operations, typically implemented pre-flight, face limitations in effectiveness due to uncertainties during airspace operations. Therefore, executing DCB during the tactical phase (as close to the departure time as possible) holds promise for better addressing these uncertainties. This study proposes a tactical-phase DCB method that accounts for uncertainties to meet practical application scenarios and requirements: compatibility with dynamic environments, high-speed computation, fairness and transparency, and high customisability. The large-scale tactical DCB problem is transformed into a hotspot-free trajectory planning problem based on sequential planning to accommodate stakeholders’ diverse performance preferences. An adaptive directed spatio-temporal graph method is introduced, enabling the integration optimisation of multiple Air Traffic Flow Management (ATFM) measures (ground delay, rerouting, and speed control) while considering flight uncertainties and fuel consumption constraints. A Heterogeneous Multi-Objective Incremental A<span><math><msup><mrow></mrow><mo>*</mo></msup></math></span> (HMOIA<span><math><msup><mrow></mrow><mo>*</mo></msup></math></span>) path search method is also developed to significantly accelerate problem-solving and meet tactical operational demands, ensuring optimal solutions by designing an admissible heuristic function. Simulation experiments based on historical European data demonstrate that the proposed method can resolve all overloaded air traffic service units with acceptable arrival delay time and fuel consumption. Compared to the Computer-Assisted Slot Allocation (CASA) method currently used in European operations, the proposed approach reduces the number of delayed flights and average delay time by approximately 79.4 % and 92.1 %, respectively. The proposed method demonstrates its value for further development to explore its potential as an upgrade to the CASA method in real-world operations.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"181 ","pages":"Article 105382"},"PeriodicalIF":7.6,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145271510","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":"Attribution explanations for decision-making in deep lane-change models","authors":"Rui Shi ,&nbsp;Tianxing Li ,&nbsp;Yasushi Yamaguchi ,&nbsp;Liguo Zhang","doi":"10.1016/j.trc.2025.105361","DOIUrl":"10.1016/j.trc.2025.105361","url":null,"abstract":"<div><div>Deep learning models are attracting considerable attention for their potential to enable intelligent lane-change behaviors. To ensure the reliability of these models, it is essential to understand their decision-making processes using attribution methods. However, existing attribution techniques, which are predominantly developed for visual tasks, often struggle to deliver accurate and interpretable explanations when applied to complex lane-change models. We identify the essential cause of this problem as the high variability in the distribution of connected perceptual information that serves as input to lane-change models. To address this, we propose a novel path aggregation attribution method, where paths describe the transition of a feature from absence to presence, expressing its relative contribution. Our method leverages an exponential family to introduce probabilistic paths and calculate attribution expectations, effectively traversing the input feature distribution space to provide a more comprehensive representation of feature transitions. Additionally, we introduce a distribution-informed counterfactual reference to define starting points of the paths, enabling the flexible generation of traffic scenarios with feature absence. Extensive experiments on three lane-change models show that our method consistently outperforms state-of-the-art attribution methods. Specifically, we achieve higher performance on four widely used quantitative metrics, <em>i.e.</em>, sensitivity-n, accuracy information curve, softmax information curve, and most-relevant-first, demonstrating superior reliability and interpretability.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"180 ","pages":"Article 105361"},"PeriodicalIF":7.6,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266542","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":"Extended visibility of autonomous vehicles via optimized cooperative perception under imperfect communication","authors":"Ahmad Sarlak ,&nbsp;Rahul Amin ,&nbsp;Abolfazl Razi","doi":"10.1016/j.trc.2025.105350","DOIUrl":"10.1016/j.trc.2025.105350","url":null,"abstract":"<div><div>Autonomous Vehicles (AVs) rely on individual perception systems to navigate safely. However, these systems face significant challenges in adverse weather conditions, complex road geometries, and dense traffic scenarios. Cooperative Perception (CP) has emerged as a promising approach to extending the perception quality of AVs by jointly processing shared camera feeds and sensor readings across multiple vehicles. This work presents a novel CP framework designed to optimize vehicle selection and networking resource utilization under imperfect communications, where message exchange between the ego and helper vehicles is subject to noise, interference, and fading, represented by nonzero packet drop rates. Our optimized CP formation considers critical factors such as the helper vehicles’ spatial position, visual range, motion blur, and available communication budgets. Furthermore, our resource optimization module allocates communication channels while adjusting power levels to maximize data flow efficiency between the ego and helper vehicles, considering realistic models of modern vehicular communication systems, such as LTE and 5G NR-V2X. We validate our approach through extensive experiments on pedestrian detection in challenging scenarios, using synthetic data generated by the CARLA simulator. The results demonstrate that our method significantly improves upon the perception quality of individual AVs with about 10 % gain in detection accuracy. This substantial gain uncovers the unleashed potential of CP to enhance AV safety and performance in complex situations.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"180 ","pages":"Article 105350"},"PeriodicalIF":7.6,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266552","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":"Joint prediction and understanding of multimodal traffic flow with a bidirectional temporal dynamic spatial hypergraph neural network model","authors":"Qin Li ,&nbsp;Zuocai Zheng ,&nbsp;Chenyang Luo ,&nbsp;Xuan Yang ,&nbsp;Yong Wang ,&nbsp;Yuankai Wu","doi":"10.1016/j.trc.2025.105358","DOIUrl":"10.1016/j.trc.2025.105358","url":null,"abstract":"<div><div>Traffic flow prediction plays a vital role in urban planning, traffic management and control. Graph convolutional networks have driven substantial advances in forecasting for a single transportation mode, yet they fall short when applied to modern multimodal networks because they do not account for interactions among coexisting modes. Although recent efforts have explored multimodal traffic prediction using multiple graph structures to extract pairwise, local spatial dependencies either within or across modes, these methods tend to lack the flexibility to capture high-order, global spatial correlations among multiple modes or functionally similar areas. In addition, multimodal traffic data often suffer from sparsity and random fluctuations caused by distributional differences, posing persistent challenges to cooperative prediction. To address these issues, this paper introduces a Bidirectional Temporal Dynamic Spatial Hypergraph Neural Network (BiT-DSHGNN). First, we construct a static hypergraph based on clusters of administrative functional areas and apply hypergraph convolution to capture intrinsic global spatial correlations among functionally related regions. We then design dynamic semantic hypergraphs that evolve over time, enabling the model to learn time-varying high-order spatial dependencies across modes through a dedicated dynamic hypergraph neural network module. This facilitates cross-modal information sharing, allowing high-density mode nodes to enrich the contextual representation of low-density nodes. Additionally, we propose a bidirectional temporal feature extraction module, named Bidirectional Temporal Gated Network (BTGN), which combines a Bidirectional Temporal Convolutional Network (BiTCN) with a Bidirectional Gated Recurrent Unit (BiGRU). This module integrates both past and future contextual information, further mitigating the impact of random fluctuations. Extensive experiments conduct on four real-world datasets (NYC-Taxi, NYC-Bike, CHI-Taxi, and CHI-Bike) demonstrate that our model consistently outperforms existing methods, achieving state-of-the-art prediction accuracy.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"180 ","pages":"Article 105358"},"PeriodicalIF":7.6,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266553","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":"Pickup and delivery problem with multi-visit drones considering soft time windows","authors":"Shanshan Meng ,&nbsp;Yanru Chen ,&nbsp;Dong Li","doi":"10.1016/j.trc.2025.105359","DOIUrl":"10.1016/j.trc.2025.105359","url":null,"abstract":"<div><div>We extend the pickup and delivery problem with combined truck–drone operation by assuming that a fleet of trucks can stop at customer nodes and launch drones to perform multiple pickup and delivery services with soft time windows. Unlike other studied pickup and delivery problems, one-to-one pickup and delivery services (in which the pickup and delivery requests have a one-to-one relationship), which are typical in instant retail and food delivery, are considered. We mathematically model a mixed-integer nonlinear program and introduce strengthening strategies to capture this scenario, with the objective of minimising the total cost, including the penalty cost of time window violations. We approximate the nonlinear power of hovering drones with a piecewise linear method and propose an efficient metaheuristic approach, along with truck waiting time optimisation, to solve large-size problems. Finally, comprehensive computational experiments are conducted, which demonstrate the applicability of the algorithm and the impacts of different configurations. The numerical results indicate the efficiency of our proposed model and the solution approach, demonstrating the potential operational gain obtained by implementing the combined system. The cost savings rate compared to the truck-only mode is more than 40% on average, and our algorithm outperforms the benchmark algorithms in the literature by more than 10% in terms of solution quality.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"180 ","pages":"Article 105359"},"PeriodicalIF":7.6,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266556","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":"Bounded-METANET: A new discrete-time second-order macroscopic traffic flow model for bounded speed","authors":"Weiming Zhao ,&nbsp;Claudio Roncoli ,&nbsp;Mehmet Yildirimoglu","doi":"10.1016/j.trc.2025.105345","DOIUrl":"10.1016/j.trc.2025.105345","url":null,"abstract":"<div><div>Macroscopic traffic flow models are essential for the analysis and control of large-scale transport networks. While second-order models like METANET capture non-equilibrium traffic dynamics, they can produce unrealistic speeds, such as negative values or those exceeding the free-flow limit, leading to unreliable simulations. This paper introduces Bounded-METANET, an enhanced second-order model designed to inherently produce physically consistent outputs. The formulation eliminates the convection term and incorporates anticipation and merging influences within the relaxation term through a mathematically bounded “virtual density” approach. Consequently, simulated speeds are confined to the range [0, <span><math><msub><mi>v</mi><mtext>free</mtext></msub></math></span>] without requiring saturation functions, improving model stability and calibration efficiency. The model was evaluated against the original METANET and the first-order Cell Transmission Model (CTM) in two case studies: one involving synthetic data from the SUMO simulator and another using empirical loop detector data from a German motorway. Bounded-METANET consistently outperformed both benchmarks by maintaining physical consistency in traffic flow dynamics. In the synthetic scenario, it achieved the lowest root mean square error for speed and density (9.97% and 17.11% reductions respectively vs. METANET), while in the real-world case it produced superior flow estimates with enhanced shockwave representation. Pareto analysis shows Bounded-METANET’s frontier dominates METANET across all speed-flow weightings. By enforcing physical bounds on traffic variables, Bounded-METANET provides a more reliable framework for traffic simulation, prediction, and control.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"180 ","pages":"Article 105345"},"PeriodicalIF":7.6,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266555","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":"Machine learning surrogates for agent-based models in transportation policy analysis","authors":"Elena Salomé Natterer ,&nbsp;Saini Rohan Rao ,&nbsp;Alejandro Tejada Lapuerta ,&nbsp;Roman Engelhardt ,&nbsp;Sebastian Hörl ,&nbsp;Klaus Bogenberger","doi":"10.1016/j.trc.2025.105360","DOIUrl":"10.1016/j.trc.2025.105360","url":null,"abstract":"<div><div>Effective traffic policies are crucial for managing congestion and reducing emissions. Agent-based transportation models (ABMs) offer a detailed analysis of how these policies affect travel behaviour at a granular level. However, computational constraints limit the number of scenarios that can be tested with ABMs and therefore their ability to find optimal policy settings.</div><div>In this proof-of-concept study, we propose a machine learning (ML)-based surrogate model to efficiently explore this vast solution space. By combining Graph Neural Networks (GNNs) with the attention mechanism from Transformers, the model predicts the effects of traffic policies on the road network at the link level.</div><div>We implement our approach in a large-scale MATSim simulation of Paris, France, covering over 30,000 road segments and 10,000 simulations, applying a policy involving capacity reduction on main roads. The ML surrogate achieves an overall <span><math><msup><mi>R</mi><mn>2</mn></msup></math></span> of 0.91; on primary roads where the policy applies, it reaches an <span><math><msup><mi>R</mi><mn>2</mn></msup></math></span> of 0.98. This study shows that the combination of GNNs and Transformer architectures can effectively serve as a surrogate for complex agent-based transportation models with the potential to enable large-scale policy optimization, helping urban planners explore a broader range of interventions more efficiently.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"180 ","pages":"Article 105360"},"PeriodicalIF":7.6,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266554","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}