Graph neural network-based deep reinforcement learning (GNN-DRL) algorithms have been widely applied in traffic signal coordination control of urban road networks. However, current traffic signal control methods based on GNN-DRL focus on integrating historical and current spatiotemporal data to represent features of traffic networks, without utilizing predicted future information to enhance traffic efficiency. This work proposes a novel spatiotemporal information-based deep reinforcement learning method for traffic signal coordination control of urban road networks. It implements a heterogeneous subgraph representation method to model spatial structures among closely related intersections, strengthening subgraph feature representations while reducing computational complexity. Additionally, a multi-scale spatiotemporal heterogeneous graph feature aggregation technique is designed. The proposed method incorporates traffic signal timing scheme, vehicle states and road network topology as graph node features. By applying a graph neural network, it captures multistep spatiotemporal information from historical, current, and predicted data, thereby enhancing network feature representation and foresight. Furthermore, a novel reward function is designed to perceive the spatiotemporal information of a road network. The function uses the betweenness centrality to evaluate the spatial importance of intersections. It introduces total number of vehicles and predicted traffic flow to dynamically assess the current traffic state and future traffic demand in the lanes. It improves the agent’s ability to perceive and use spatiotemporal information to make decisions. We evaluated our proposed method through experiments under three different traffic scenarios: low, medium, and high flows. The results clearly demonstrate that the proposed method outperforms existing state-of-the-art methods, by reducing average queue length by 34.12%-59.45%, maximum queue length by 25.31%-47.83%, lane occupancy rates by 27.22%-51.56%, and vehicle count by 27.29%-51.92%. Meanwhile, experiments on computational overhead and real-road networks further confirm that SIDRL offers advantages in terms of low cost and high performance. This presents new technical insights for the real-time deployment and resource optimization of urban traffic signal control.
{"title":"Deep Reinforcement Learning Based on Spatiotemporal Information for Network-Wide Traffic Signal Coordination Control","authors":"Bao-Lin Ye;Peng Wu;Lingxi Li;Weimin Wu;Bo Song;Xianchao Zhang","doi":"10.1109/OJITS.2025.3627135","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3627135","url":null,"abstract":"Graph neural network-based deep reinforcement learning (GNN-DRL) algorithms have been widely applied in traffic signal coordination control of urban road networks. However, current traffic signal control methods based on GNN-DRL focus on integrating historical and current spatiotemporal data to represent features of traffic networks, without utilizing predicted future information to enhance traffic efficiency. This work proposes a novel spatiotemporal information-based deep reinforcement learning method for traffic signal coordination control of urban road networks. It implements a heterogeneous subgraph representation method to model spatial structures among closely related intersections, strengthening subgraph feature representations while reducing computational complexity. Additionally, a multi-scale spatiotemporal heterogeneous graph feature aggregation technique is designed. The proposed method incorporates traffic signal timing scheme, vehicle states and road network topology as graph node features. By applying a graph neural network, it captures multistep spatiotemporal information from historical, current, and predicted data, thereby enhancing network feature representation and foresight. Furthermore, a novel reward function is designed to perceive the spatiotemporal information of a road network. The function uses the betweenness centrality to evaluate the spatial importance of intersections. It introduces total number of vehicles and predicted traffic flow to dynamically assess the current traffic state and future traffic demand in the lanes. It improves the agent’s ability to perceive and use spatiotemporal information to make decisions. We evaluated our proposed method through experiments under three different traffic scenarios: low, medium, and high flows. The results clearly demonstrate that the proposed method outperforms existing state-of-the-art methods, by reducing average queue length by 34.12%-59.45%, maximum queue length by 25.31%-47.83%, lane occupancy rates by 27.22%-51.56%, and vehicle count by 27.29%-51.92%. Meanwhile, experiments on computational overhead and real-road networks further confirm that SIDRL offers advantages in terms of low cost and high performance. This presents new technical insights for the real-time deployment and resource optimization of urban traffic signal control.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1424-1438"},"PeriodicalIF":5.3,"publicationDate":"2025-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11222077","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145455750","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-20DOI: 10.1109/OJITS.2025.3623131
Long Qin;Yi Shi;Xin Zhang;Peichun Liao;Yongjie Li;Xianshi Zhang;Hongmei Yan
The perception of night scenes is of crucial importance for driving safety. In the dimly lit night environment, as the visibility of objects decreases, both experienced and inexperienced drivers often struggle to fully notice the objects closely related to the driving task. Moreover, because the contours of many objects are blurred in dim night, locating and detecting objects are much more difficult than that in daytime scenes, especially for the small traffic objects, which undoubtedly greatly increases the potential road hazards. Till now, there are few studies specifically focusing on the night object detection based on driver’s attention. This research is dedicated to solving the detection problem of significant objects in night scenes, particularly small salient objects. First, we constructed a Night Eye-Tracking Object Detection Dataset (NETOD), which can provide a benchmark for research on attention-driven object detection in night scenes. Then, we proposed a salient object detection model for night traffic scenes, named NS-YOLO. NS-YOLO integrates a Bio-Inspired Spotlight Attention Module (BSAM) that combines bottom-up feature enhancement with top-down semantic guidance to accurately localize salient objects. Additionally, a hierarchical multi-scale detection architecture is introduced, leveraging cross-layer feature pyramid and dynamic upsampling to enhance the detection of small objects. The experimental results on the NETOD dataset show that the proposed salient small object detection model for night traffic scenes achieved mean Average Precision (mAP) value of 93.0%, outperforming other advanced models. It has important potential application values in driver assistance, danger warning, and other aspects, and is expected to significantly improve the safety and intelligence of night driving. Beyond technical advancements, this work highlights the necessity of human-centric attention mechanisms in autonomous systems, paving the way for safer and more interpretable AI-driven vehicles.
{"title":"Salient Object Detection of Dynamic Night Scenes via Bio-Inspired Spotlight Attention and Hierarchical Edge-Texture Fusion","authors":"Long Qin;Yi Shi;Xin Zhang;Peichun Liao;Yongjie Li;Xianshi Zhang;Hongmei Yan","doi":"10.1109/OJITS.2025.3623131","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3623131","url":null,"abstract":"The perception of night scenes is of crucial importance for driving safety. In the dimly lit night environment, as the visibility of objects decreases, both experienced and inexperienced drivers often struggle to fully notice the objects closely related to the driving task. Moreover, because the contours of many objects are blurred in dim night, locating and detecting objects are much more difficult than that in daytime scenes, especially for the small traffic objects, which undoubtedly greatly increases the potential road hazards. Till now, there are few studies specifically focusing on the night object detection based on driver’s attention. This research is dedicated to solving the detection problem of significant objects in night scenes, particularly small salient objects. First, we constructed a Night Eye-Tracking Object Detection Dataset (NETOD), which can provide a benchmark for research on attention-driven object detection in night scenes. Then, we proposed a salient object detection model for night traffic scenes, named NS-YOLO. NS-YOLO integrates a Bio-Inspired Spotlight Attention Module (BSAM) that combines bottom-up feature enhancement with top-down semantic guidance to accurately localize salient objects. Additionally, a hierarchical multi-scale detection architecture is introduced, leveraging cross-layer feature pyramid and dynamic upsampling to enhance the detection of small objects. The experimental results on the NETOD dataset show that the proposed salient small object detection model for night traffic scenes achieved mean Average Precision (mAP) value of 93.0%, outperforming other advanced models. It has important potential application values in driver assistance, danger warning, and other aspects, and is expected to significantly improve the safety and intelligence of night driving. Beyond technical advancements, this work highlights the necessity of human-centric attention mechanisms in autonomous systems, paving the way for safer and more interpretable AI-driven vehicles.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1377-1390"},"PeriodicalIF":5.3,"publicationDate":"2025-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11207722","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145351957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The rectifier traction substations used in subway power supply systems cannot feed back regenerative energy (RE), leading to high energy consumption. The subway power supply system is retrofitted to effectively feed back and utilize RE based on bidirectional converters (BC) to achieve real-time energy sharing in traction substations. This study conducted a collaborative optimization of the site and characteristics of adding BC to subway substations, reaching the lowest economic cost for the subway system. First, a power flow calculation platform compatible with different types and parameters of substations is established to calculate the parameters of the subway power supply and train operation processes. Then, a minimum-cost objective optimization model is designed, which comprises the equipment and power consumption costs associated with adding BC. A dual-population fusion algorithm based on the red-billed blue magpie optimizer algorithm and rime algorithm is proposed to obtain BCs’ optimal site and characteristics scheme. Finally, the effectiveness of the model is verified through a detailed case study, showing that the comprehensive cost can be reduced by 22.32% compared to traditional rectifier traction substations, and the voltage fluctuation of the overhead contact line is effectively suppressed, ensuring the economical and reliable operation of the subway power supply system.
{"title":"Site and Characteristic Optimization of Retrofitted Bidirectional Converters in Subway Traction Substations Considering Integrated Cost","authors":"Chengcheng Fu;Pengfei Sun;Qingyuan Wang;Xiaoyun Feng","doi":"10.1109/OJITS.2025.3621849","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3621849","url":null,"abstract":"The rectifier traction substations used in subway power supply systems cannot feed back regenerative energy (RE), leading to high energy consumption. The subway power supply system is retrofitted to effectively feed back and utilize RE based on bidirectional converters (BC) to achieve real-time energy sharing in traction substations. This study conducted a collaborative optimization of the site and characteristics of adding BC to subway substations, reaching the lowest economic cost for the subway system. First, a power flow calculation platform compatible with different types and parameters of substations is established to calculate the parameters of the subway power supply and train operation processes. Then, a minimum-cost objective optimization model is designed, which comprises the equipment and power consumption costs associated with adding BC. A dual-population fusion algorithm based on the red-billed blue magpie optimizer algorithm and rime algorithm is proposed to obtain BCs’ optimal site and characteristics scheme. Finally, the effectiveness of the model is verified through a detailed case study, showing that the comprehensive cost can be reduced by 22.32% compared to traditional rectifier traction substations, and the voltage fluctuation of the overhead contact line is effectively suppressed, ensuring the economical and reliable operation of the subway power supply system.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1412-1423"},"PeriodicalIF":5.3,"publicationDate":"2025-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11204498","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145405342","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-14DOI: 10.1109/OJITS.2025.3621578
Tex Ruskamp;Marta Ribeiro;Ferdinand Dijkstra
Reducing uncertainty in air traffic flow management is crucial for maintaining safety and efficiency in modern aviation. In particular, forecasting Actual Take-Off Times (ATOT) for flights across Europe is challenging due to the diverse flight-specific variables and operational conditions. Additionally, to help operations, this prediction must be done well in advance in order to prevent future traffic densities from being higher than the airspace capacity. However, recent studies often make predictions on shorter horizons and do not consider the effect of knock-on delays. This study covers this gap, by focusing on larger prediction horizons and different types of delay. We enhance ATOT prediction for flights arriving at Amsterdam Schiphol Airport from European out-stations by leveraging machine learning techniques, specifically a Long Short-Term Memory (LSTM) neural network, augmented with a Multihead Attention mechanism. A model capable of capturing complex temporal dependencies and operational factors influencing the ATOT is developed utilizing data from Electronic Flight Data messages, weather reports and a EUROCONTROL dataset. The model’s performance is evaluated against traditional ensemble methods and the current Decision Support Tool (DST) system used by Luchtverkeersleiding Nederland (LVNL). Results indicate that the LSTM model outperforms existing models including a reproduction of the DST, achieving a Mean Absolute Error of 12.05 minutes at a forecast horizon of 4 hours, demonstrating significant improvements. Finally, this assessment underscores the importance of factors such as the knock-on effect in delay prediction can significantly enhance demand forecasting, leading to more efficient air traffic management and reduced delays.
{"title":"Prediction of Traffic Take-Off Times at Out-Stations: A Case Study at Schiphol Airport","authors":"Tex Ruskamp;Marta Ribeiro;Ferdinand Dijkstra","doi":"10.1109/OJITS.2025.3621578","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3621578","url":null,"abstract":"Reducing uncertainty in air traffic flow management is crucial for maintaining safety and efficiency in modern aviation. In particular, forecasting Actual Take-Off Times (ATOT) for flights across Europe is challenging due to the diverse flight-specific variables and operational conditions. Additionally, to help operations, this prediction must be done well in advance in order to prevent future traffic densities from being higher than the airspace capacity. However, recent studies often make predictions on shorter horizons and do not consider the effect of knock-on delays. This study covers this gap, by focusing on larger prediction horizons and different types of delay. We enhance ATOT prediction for flights arriving at Amsterdam Schiphol Airport from European out-stations by leveraging machine learning techniques, specifically a Long Short-Term Memory (LSTM) neural network, augmented with a Multihead Attention mechanism. A model capable of capturing complex temporal dependencies and operational factors influencing the ATOT is developed utilizing data from Electronic Flight Data messages, weather reports and a EUROCONTROL dataset. The model’s performance is evaluated against traditional ensemble methods and the current Decision Support Tool (DST) system used by Luchtverkeersleiding Nederland (LVNL). Results indicate that the LSTM model outperforms existing models including a reproduction of the DST, achieving a Mean Absolute Error of 12.05 minutes at a forecast horizon of 4 hours, demonstrating significant improvements. Finally, this assessment underscores the importance of factors such as the knock-on effect in delay prediction can significantly enhance demand forecasting, leading to more efficient air traffic management and reduced delays.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1391-1411"},"PeriodicalIF":5.3,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11202962","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145352012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-26DOI: 10.1109/OJITS.2025.3614862
Chinthaka Premachandra;Eigo Ito
In today’s motorized society, road accidents occur frequently, and their incidence continues to rise with the increasing number of car users worldwide. A significant proportion of these accidents occur at intersections, where one promising countermeasure is the use of multi-camera systems that assist pedestrians and drivers by detecting moving vehicles in the intersection area. However, conventional vehicle detection methods suffer from reduced accuracy as vehicles move farther from the camera, since distant vehicles appear smaller in images. To address this limitation, we propose a method that first identifies the distant region of the road in an image and then applies up-sampling to enhance the visibility of faraway road area for improved vehicle detection. In the proposed approach, nearby moving vehicles are roughly extracted using inter-frame subtraction across consecutive frames, and these subtractions are accumulated over time as trajectories. Based on these trajectories, we introduce a novel method to estimate the road’s vanishing point, which is then used to determine the distant road area. This region is subsequently up-sampled in consecutive frames, and vehicle detection is performed using a Gaussian Mixture Model (GMM) to identify distant vehicles. Extensive experiments confirm the effectiveness of the proposed method. The results demonstrated that, although detection accuracy naturally decreases with distance, our method achieves more than twice the accuracy of conventional approaches under both daytime and nighttime conditions.
{"title":"Recognizing Distant Vehicles on GMM by Extracting Far Road Area Based on Analyzing Trajectories of Nearby Vehicles","authors":"Chinthaka Premachandra;Eigo Ito","doi":"10.1109/OJITS.2025.3614862","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3614862","url":null,"abstract":"In today’s motorized society, road accidents occur frequently, and their incidence continues to rise with the increasing number of car users worldwide. A significant proportion of these accidents occur at intersections, where one promising countermeasure is the use of multi-camera systems that assist pedestrians and drivers by detecting moving vehicles in the intersection area. However, conventional vehicle detection methods suffer from reduced accuracy as vehicles move farther from the camera, since distant vehicles appear smaller in images. To address this limitation, we propose a method that first identifies the distant region of the road in an image and then applies up-sampling to enhance the visibility of faraway road area for improved vehicle detection. In the proposed approach, nearby moving vehicles are roughly extracted using inter-frame subtraction across consecutive frames, and these subtractions are accumulated over time as trajectories. Based on these trajectories, we introduce a novel method to estimate the road’s vanishing point, which is then used to determine the distant road area. This region is subsequently up-sampled in consecutive frames, and vehicle detection is performed using a Gaussian Mixture Model (GMM) to identify distant vehicles. Extensive experiments confirm the effectiveness of the proposed method. The results demonstrated that, although detection accuracy naturally decreases with distance, our method achieves more than twice the accuracy of conventional approaches under both daytime and nighttime conditions.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1346-1357"},"PeriodicalIF":5.3,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11182307","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145255924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The increase in vehicle density exacerbates traffic congestion, accidents, and emissions. Automated Vehicles (AVs), while promising improved safety and efficiency, require seamless coordination and communication to unlock their full potential. The European Telecommunications Standards Institute (ETSI) Maneuver Coordination Service (MCS) draft introduces Vehicle-to-Everything (V2X) communication for real-time vehicle coordination, utilizing a modular architecture designed to enhance inter-vehicle communication. However, a major limitation of the current MCS framework is its vulnerability to message loss during maneuver negotiation, which can increase latency and negatively impact maneuver efficiency. This paper proposes an acknowledgment mechanism in MCS to enhance message reliability and a Relevance Message Detector to filter out irrelevant messages, reducing processing overhead. The experimental results demonstrate that introducing an acknowledgment mechanism can reduce maneuver negotiation time by approximately 900 ms compared to standard methods under packet loss scenarios, significantly improving reliability and efficiency. Furthermore, the Relevance Message Detector effectively minimizes unnecessary message processing, enhancing overall system efficiency. Functional evaluations validate the correct execution of coordinated maneuvers, demonstrating the practical benefits of the proposed extensions. These enhancements contribute to a more robust and efficient MCS framework, improving AV coordination in real-world scenarios.
{"title":"Maneuver Coordination Service With Reliability and Relevance Enhancements","authors":"Andreia Figueiredo;João Viegas;Pedro Rito;Miguel Luís;Susana Sargento","doi":"10.1109/OJITS.2025.3613990","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3613990","url":null,"abstract":"The increase in vehicle density exacerbates traffic congestion, accidents, and emissions. Automated Vehicles (AVs), while promising improved safety and efficiency, require seamless coordination and communication to unlock their full potential. The European Telecommunications Standards Institute (ETSI) Maneuver Coordination Service (MCS) draft introduces Vehicle-to-Everything (V2X) communication for real-time vehicle coordination, utilizing a modular architecture designed to enhance inter-vehicle communication. However, a major limitation of the current MCS framework is its vulnerability to message loss during maneuver negotiation, which can increase latency and negatively impact maneuver efficiency. This paper proposes an acknowledgment mechanism in MCS to enhance message reliability and a Relevance Message Detector to filter out irrelevant messages, reducing processing overhead. The experimental results demonstrate that introducing an acknowledgment mechanism can reduce maneuver negotiation time by approximately 900 ms compared to standard methods under packet loss scenarios, significantly improving reliability and efficiency. Furthermore, the Relevance Message Detector effectively minimizes unnecessary message processing, enhancing overall system efficiency. Functional evaluations validate the correct execution of coordinated maneuvers, demonstrating the practical benefits of the proposed extensions. These enhancements contribute to a more robust and efficient MCS framework, improving AV coordination in real-world scenarios.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1325-1345"},"PeriodicalIF":5.3,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11177009","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145255974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-22DOI: 10.1109/OJITS.2025.3613259
Mohd A. Khan;Wilco Burghout;Oded Cats;Erik Jenelius;Matej Cebecauer
Recent advances in automation have accelerated the development of autonomous electric vehicles (AEVs), which offer the potential for continuous operation, constrained primarily by the need for recharging. We propose a dynamic charging strategy based on Mobile Autonomous Charging Pods (MAPs), which are battery-equipped electric vehicles capable of transferring energy to AEVs while in motion. We introduce a dedicated simulation framework within the microscopic traffic simulator SUMO, incorporating MAP-specific modules for assignment, navigation, and real-time energy transfer under realistic traffic constraints. We model the behavior of both MAPs and AEVs in a stylized looped network and evaluate system-level performance under various demand and fleet configurations. Key performance indicators include energy consumption, charging efficiency, battery utilization, and reductions in AEV battery capacity requirements. Simulation results demonstrate that MAPs can effectively support continuous AEV operation, achieving up to 14% battery downsizing with minimal infrastructure investment, while also reducing travel time by 7%, relative to fixed charging solutions. This study lays the foundation for simulation-based evaluation of MAP-based dynamic charging as a scalable, flexible, and efficient alternative to fixed charging solutions.
{"title":"A Simulation Framework for Evaluating Mobile Autonomous Charging Pod Operations","authors":"Mohd A. Khan;Wilco Burghout;Oded Cats;Erik Jenelius;Matej Cebecauer","doi":"10.1109/OJITS.2025.3613259","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3613259","url":null,"abstract":"Recent advances in automation have accelerated the development of autonomous electric vehicles (AEVs), which offer the potential for continuous operation, constrained primarily by the need for recharging. We propose a dynamic charging strategy based on Mobile Autonomous Charging Pods (MAPs), which are battery-equipped electric vehicles capable of transferring energy to AEVs while in motion. We introduce a dedicated simulation framework within the microscopic traffic simulator SUMO, incorporating MAP-specific modules for assignment, navigation, and real-time energy transfer under realistic traffic constraints. We model the behavior of both MAPs and AEVs in a stylized looped network and evaluate system-level performance under various demand and fleet configurations. Key performance indicators include energy consumption, charging efficiency, battery utilization, and reductions in AEV battery capacity requirements. Simulation results demonstrate that MAPs can effectively support continuous AEV operation, achieving up to 14% battery downsizing with minimal infrastructure investment, while also reducing travel time by 7%, relative to fixed charging solutions. This study lays the foundation for simulation-based evaluation of MAP-based dynamic charging as a scalable, flexible, and efficient alternative to fixed charging solutions.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1282-1297"},"PeriodicalIF":5.3,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11175572","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145255929","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-17DOI: 10.1109/OJITS.2025.3610928
Huamin Li;Moye Lu;Junfeng Mao;Xiaojun Yu
This study addresses the challenge of bridging macroscopic optimization and microscopic driving behavior under communication uncertainty in Intelligent Vehicle Cyber-Physical Systems (IVCPS). A multi-objective macroscopic optimization model is first developed to generate recommended speeds, with different evolutionary algorithms systematically compared. Through experiments with Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), and Real-Coded Genetic Algorithm (RCGA), RCGA is identified as the most effective solver. The recommended speeds are subsequently integrated into the microscopic layer, where a modified Intelligent Driver Model (IDM) accounts for both multi-preceding vehicle interactions and macroscopic guidance. Communication uncertainty in the transmission process is modeled and quantified using soft set theory, enabling robust adaptation of vehicle behaviors. Simulation results under both ideal and uncertain communication conditions demonstrate that: (i) the proposed framework consistently outperforms the baseline IDM and the conventional IDM with recommended speeds, validating its effectiveness; (ii) variations in optimization weights significantly influence the performance of the modified IDM; and (iii) the modified IDM achieves superior traffic efficiency and fuel economy across different traffic demand scenarios. Overall, the findings highlight the necessity of incorporating uncertainty-aware speed guidance to effectively link macroscopic optimization with microscopic control, offering new insights into building resilient and efficient intelligent transportation systems.
{"title":"The Interaction of Macroscopic Optimization and Microscopic Traffic Flow With Communication Uncertainty in Intelligent Vehicle Cyber–Physical System","authors":"Huamin Li;Moye Lu;Junfeng Mao;Xiaojun Yu","doi":"10.1109/OJITS.2025.3610928","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3610928","url":null,"abstract":"This study addresses the challenge of bridging macroscopic optimization and microscopic driving behavior under communication uncertainty in Intelligent Vehicle Cyber-Physical Systems (IVCPS). A multi-objective macroscopic optimization model is first developed to generate recommended speeds, with different evolutionary algorithms systematically compared. Through experiments with Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), and Real-Coded Genetic Algorithm (RCGA), RCGA is identified as the most effective solver. The recommended speeds are subsequently integrated into the microscopic layer, where a modified Intelligent Driver Model (IDM) accounts for both multi-preceding vehicle interactions and macroscopic guidance. Communication uncertainty in the transmission process is modeled and quantified using soft set theory, enabling robust adaptation of vehicle behaviors. Simulation results under both ideal and uncertain communication conditions demonstrate that: (i) the proposed framework consistently outperforms the baseline IDM and the conventional IDM with recommended speeds, validating its effectiveness; (ii) variations in optimization weights significantly influence the performance of the modified IDM; and (iii) the modified IDM achieves superior traffic efficiency and fuel economy across different traffic demand scenarios. Overall, the findings highlight the necessity of incorporating uncertainty-aware speed guidance to effectively link macroscopic optimization with microscopic control, offering new insights into building resilient and efficient intelligent transportation systems.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1265-1281"},"PeriodicalIF":5.3,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11168868","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145210119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-16DOI: 10.1109/OJITS.2025.3610456
Shakib Mustavee;Shaurya Agarwal
This paper presents a Koopman operator-based approach for the car-following model using SwarmDMD, a dynamic mode decomposition (DMD)-type algorithm designed to capture multi-agent interactions. A central challenge in Koopman operator-based car-following dynamics modeling lies in selecting an appropriate dictionary of observable functions. While previous studies have demonstrated various techniques, including deep learning, to learn the Koopman operator, they do not yield analytical forms. To address this, we revisit classical physics-based car-following models and propose candidate observables inspired by their mathematical structures. These observables are used within the Koopman and DMD framework to reconstruct a follower’s acceleration. The corresponding speed and trajectory are then estimated from the reconstructed acceleration. We evaluate the framework using both simulated and real-world datasets, demonstrating strong potential for accuracy and interpretability. While this study focuses on single-lane human-driven vehicles (HDVs), the framework is easily extendable to multi-lane traffic and connected and autonomous vehicle (CAV) scenarios, highlighting its generality and versatility. We presented a comparative evaluation of the proposed model by contrasting its acceleration reconstruction performance with that of both physics-based and data-driven models. Additionally, we interpreted the individual entries of the SwarmDMD matrix by establishing their connections to parameters of physics-based models. The codes and data used in the paper are available at our GitHub page.
{"title":"A Koopman-Theoretic Approach to Car-Following and Multi-Lane Interaction Modeling","authors":"Shakib Mustavee;Shaurya Agarwal","doi":"10.1109/OJITS.2025.3610456","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3610456","url":null,"abstract":"This paper presents a Koopman operator-based approach for the car-following model using SwarmDMD, a dynamic mode decomposition (DMD)-type algorithm designed to capture multi-agent interactions. A central challenge in Koopman operator-based car-following dynamics modeling lies in selecting an appropriate dictionary of observable functions. While previous studies have demonstrated various techniques, including deep learning, to learn the Koopman operator, they do not yield analytical forms. To address this, we revisit classical physics-based car-following models and propose candidate observables inspired by their mathematical structures. These observables are used within the Koopman and DMD framework to reconstruct a follower’s acceleration. The corresponding speed and trajectory are then estimated from the reconstructed acceleration. We evaluate the framework using both simulated and real-world datasets, demonstrating strong potential for accuracy and interpretability. While this study focuses on single-lane human-driven vehicles (HDVs), the framework is easily extendable to multi-lane traffic and connected and autonomous vehicle (CAV) scenarios, highlighting its generality and versatility. We presented a comparative evaluation of the proposed model by contrasting its acceleration reconstruction performance with that of both physics-based and data-driven models. Additionally, we interpreted the individual entries of the SwarmDMD matrix by establishing their connections to parameters of physics-based models. The codes and data used in the paper are available at our GitHub page.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1358-1376"},"PeriodicalIF":5.3,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11165106","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145255946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Multimodal Intelligent Transportation Systems (M-ITS) encompass a range of transportation services utilizing various modes of transport (e.g., buses, trains, ride-sharing) and incorporating intelligent technologies for enhanced efficiency and user experience. Traditional, non-adaptive system architectures struggle to respond to dynamic changes in real-time traffic conditions, user demands, and operational disruptions. These rigid systems lack flexibility in integrating new technologies, managing fluctuating demand, and ensuring seamless operation across multiple transport modes. Consequently, inefficiencies in data handling, scalability, and real-time decision-making emerge, hindering the potential of M-ITS. In this paper, we provide a conceptual layered architecture that can adapt to various needs of multimodal transportation systems. The proposed architecture focuses on aspects such as scalability, adaptability, seamless integration, and interoperability of various subcomponents that are owned and managed by different stakeholders (parties with an interest or role in the system, such as users, city planners, service operators, and technology providers). In addition to the component architecture, we propose a data architecture that emphasizes the crucial role of integrating multimodal, multisource data to enable intelligent decision-making. We illustrate the functionality of the proposed architecture through two use cases at a conceptual level: a traffic monitoring system and a traffic flow prediction system. These examples demonstrate how the data and system architecture can be fused and serve multimodal intelligent transport services, highlighting its ability to adapt to complex urban environments. Furthermore, we present results for an emergency vehicle approaching scenario, showcasing the architecture’s responsiveness and adaptability in critical situations.
{"title":"Adaptive System Architecture for Intelligent Multimodal Transport: Challenges and Fundamental Design Aspects","authors":"Fatemeh Golpayegani;Abdollah Malekjafarian;Muhammad Farooq;Saeedeh Ghanadbashi;Nima Afraz","doi":"10.1109/OJITS.2025.3609482","DOIUrl":"https://doi.org/10.1109/OJITS.2025.3609482","url":null,"abstract":"Multimodal Intelligent Transportation Systems (M-ITS) encompass a range of transportation services utilizing various modes of transport (e.g., buses, trains, ride-sharing) and incorporating intelligent technologies for enhanced efficiency and user experience. Traditional, non-adaptive system architectures struggle to respond to dynamic changes in real-time traffic conditions, user demands, and operational disruptions. These rigid systems lack flexibility in integrating new technologies, managing fluctuating demand, and ensuring seamless operation across multiple transport modes. Consequently, inefficiencies in data handling, scalability, and real-time decision-making emerge, hindering the potential of M-ITS. In this paper, we provide a conceptual layered architecture that can adapt to various needs of multimodal transportation systems. The proposed architecture focuses on aspects such as scalability, adaptability, seamless integration, and interoperability of various subcomponents that are owned and managed by different stakeholders (parties with an interest or role in the system, such as users, city planners, service operators, and technology providers). In addition to the component architecture, we propose a data architecture that emphasizes the crucial role of integrating multimodal, multisource data to enable intelligent decision-making. We illustrate the functionality of the proposed architecture through two use cases at a conceptual level: a traffic monitoring system and a traffic flow prediction system. These examples demonstrate how the data and system architecture can be fused and serve multimodal intelligent transport services, highlighting its ability to adapt to complex urban environments. Furthermore, we present results for an emergency vehicle approaching scenario, showcasing the architecture’s responsiveness and adaptability in critical situations.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"6 ","pages":"1298-1324"},"PeriodicalIF":5.3,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11160689","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145255930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：