Pub Date : 2025-06-04DOI: 10.1109/TIV.2024.3496635
{"title":"TechRxiv: Share Your Preprint Research with the World!","authors":"","doi":"10.1109/TIV.2024.3496635","DOIUrl":"https://doi.org/10.1109/TIV.2024.3496635","url":null,"abstract":"","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"9 9","pages":"5970-5970"},"PeriodicalIF":14.0,"publicationDate":"2025-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11023922","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144213674","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-23DOI: 10.1109/TIV.2024.3519366
Bolin Liao;Tinglei Wang;Xinwei Cao;Cheng Hua;Shuai Li
In multi-agent real-time position management tasks, the accuracy of error convergence and convergence time are crucial. This paper reformulates the proposed real-time position management scheme as a quadratic programming problem with equality constraints and solves it in real-time using the zeroing neural dynamics (ZND) model. To enhance the model's ability to detect real-time position management errors, an adaptive parameter finite-time convergent zeroing neural dynamics (AP-FTZND) model is introduced, incorporating adaptive parameters and a nonlinear activation function (AF) within the ZND framework. The global convergence of the AP-FTZND model is proven using the Lyapunov theory, and the upper bound of the convergence time is derived. Finally, the effectiveness and superiority of the AP-FTZND model in solving multi-agent real-time position management tasks are validated through simulations and physical experiments.
{"title":"Novel Zeroing Neural Dynamics for Real-Time Management of Multi-Vehicle Cooperation","authors":"Bolin Liao;Tinglei Wang;Xinwei Cao;Cheng Hua;Shuai Li","doi":"10.1109/TIV.2024.3519366","DOIUrl":"https://doi.org/10.1109/TIV.2024.3519366","url":null,"abstract":"In multi-agent real-time position management tasks, the accuracy of error convergence and convergence time are crucial. This paper reformulates the proposed real-time position management scheme as a quadratic programming problem with equality constraints and solves it in real-time using the zeroing neural dynamics (ZND) model. To enhance the model's ability to detect real-time position management errors, an adaptive parameter finite-time convergent zeroing neural dynamics (AP-FTZND) model is introduced, incorporating adaptive parameters and a nonlinear activation function (AF) within the ZND framework. The global convergence of the AP-FTZND model is proven using the Lyapunov theory, and the upper bound of the convergence time is derived. Finally, the effectiveness and superiority of the AP-FTZND model in solving multi-agent real-time position management tasks are validated through simulations and physical experiments.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 12","pages":"5197-5212"},"PeriodicalIF":14.3,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10812069","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145772061","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Autonomous Vehicles face significant safety challenges in complex urban environments, particularly in detecting and tracking vulnerable road users like pedestrians and cyclists, who are at higher risk of fatal accidents. This paper explores the potential of Ultra-Wideband technology as an additional sensing modality, known for its high ranging accuracy and robustness in challenging environments. Through real-world experiments, we provide a qualitative analysis of Ultra-Wideband performance in scenarios prone to intermittent vision failures, demonstrating its effectiveness in improving vulnerable road users' detection in urban driving scenarios. To enable its widespread application in autonomous driving, we also present WiDEVIEW, the first multimodal dataset that integrates LiDAR, three RGB cameras, GPS/IMU, and Ultra-Wideband sensors for providing urban driving scenarios with extensive pedestrian-vehicle interactions, which can aid in studying pedestrian-vehicle interactions, developing better pedestrian detection and tracking and eventually safe autonomous navigation algorithms by augmenting Ultra-Wideband and using the complimentary properties of Ultra-Wideband sensing with vision and LiDAR data. Finally, we demonstrate the potential applications of the Ultra-Wideband technology in vehicle to vehicle communication and vulnerable road users localization scenarios.
{"title":"Ultra-Wideband Technology for Improved Detection of Vulnerable Road Users in Urban Settings: Dataset and Evaluation","authors":"Jia Huang;Alvika Gautam;Junghun Choi;Srikanth Saripalli","doi":"10.1109/TIV.2024.3521215","DOIUrl":"https://doi.org/10.1109/TIV.2024.3521215","url":null,"abstract":"Autonomous Vehicles face significant safety challenges in complex urban environments, particularly in detecting and tracking vulnerable road users like pedestrians and cyclists, who are at higher risk of fatal accidents. This paper explores the potential of Ultra-Wideband technology as an additional sensing modality, known for its high ranging accuracy and robustness in challenging environments. Through real-world experiments, we provide a qualitative analysis of Ultra-Wideband performance in scenarios prone to intermittent vision failures, demonstrating its effectiveness in improving vulnerable road users' detection in urban driving scenarios. To enable its widespread application in autonomous driving, we also present WiDEVIEW, the first multimodal dataset that integrates LiDAR, three RGB cameras, GPS/IMU, and Ultra-Wideband sensors for providing urban driving scenarios with extensive pedestrian-vehicle interactions, which can aid in studying pedestrian-vehicle interactions, developing better pedestrian detection and tracking and eventually safe autonomous navigation algorithms by augmenting Ultra-Wideband and using the complimentary properties of Ultra-Wideband sensing with vision and LiDAR data. Finally, we demonstrate the potential applications of the Ultra-Wideband technology in vehicle to vehicle communication and vulnerable road users localization scenarios.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 12","pages":"5278-5287"},"PeriodicalIF":14.3,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145772101","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 : 2024-12-18DOI: 10.1109/TIV.2024.3520199
Xiaozheng Jin;Yuhan Hou
This paper is dedicated to address the event-triggered finite-time trajectory tracking control problem of a category of perturbed quadrotor autonomous aerial vehicles (AAVs) against partially norm-bounded and state-dependent perturbations. Adaptive compensation control strategies are developed to attenuate perturbation-induced impacts on the position and attitude tracking subsystems of the quadrotor AAVs. To minimize communication resources and actuation of actuators, event-triggered control techniques with novel triggering mechanisms are introduced to adapt control inputs based on the developed adaptive finite-time compensation control strategies. For the purpose of guaranteeing the absence of Zeno phenomenon in the quadrotor AAVs, bounds on the inter-event execution time are defined. By incorporating the event-triggering conditions, bounded and asymptotic trajectory tracking results of position and attitude AAV subsystems are achieved respectively by utilizing Lyapunov stability theory under the influence of general perturbations. The effectiveness of the presented control strategies is validated through comparative simulation results of a quadrotor AAV system.
{"title":"Adaptive Event-Triggered Finite-Time Tracking Control for a Class of Perturbed Quadrotor Autonomous Aerial Vehicles","authors":"Xiaozheng Jin;Yuhan Hou","doi":"10.1109/TIV.2024.3520199","DOIUrl":"https://doi.org/10.1109/TIV.2024.3520199","url":null,"abstract":"This paper is dedicated to address the event-triggered finite-time trajectory tracking control problem of a category of perturbed quadrotor autonomous aerial vehicles (AAVs) against partially norm-bounded and state-dependent perturbations. Adaptive compensation control strategies are developed to attenuate perturbation-induced impacts on the position and attitude tracking subsystems of the quadrotor AAVs. To minimize communication resources and actuation of actuators, event-triggered control techniques with novel triggering mechanisms are introduced to adapt control inputs based on the developed adaptive finite-time compensation control strategies. For the purpose of guaranteeing the absence of Zeno phenomenon in the quadrotor AAVs, bounds on the inter-event execution time are defined. By incorporating the event-triggering conditions, bounded and asymptotic trajectory tracking results of position and attitude AAV subsystems are achieved respectively by utilizing Lyapunov stability theory under the influence of general perturbations. The effectiveness of the presented control strategies is validated through comparative simulation results of a quadrotor AAV system.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 12","pages":"5262-5277"},"PeriodicalIF":14.3,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145772059","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}
The rapid adoption of Electric Vehicles (EVs) in the global pursuit of energy efficiency and carbon neutrality necessitates effective strategies to mitigate their carbon footprint and enhance operational stability. Similarly, in order to achieve Sustainability Development Goals, a promising solution toward green mobility, which is gaining ground nowadays, constitutes Automated Vehicles (AVs), which are EVs having the capability to move autonomously, without the need for a driver. One of the most critical factors regarding energy efficiency is the optimal management of energy consumption of AVs. This research study explores the application of machine learning (ML) models for State-of-Charge (SoC) forecasting in AVs, crucial for addressing challenges such as range anxiety and grid overloading. Leveraging real-life EV data from automated minibuses in Gothenburg, Sweeden, a comprehensive pipeline is proposed for data pre-processing, feature selection, and model training. With a focus on predicting SoC several minutes ahead, various ML techniques, including linear regression, ridge regression, lasso regression, and elastic-net regression are embedded in a pipeline specifically developed to overcome the challenge of training time-series models on discontinuous data segments, corresponding to discharge cycles. This pipeline is called Cross-Segment-Leakage-Free (CSLF). The results demonstrate the efficacy of CSLF, with the best-performing model achieving a Mean Absolute Error (MAE) of 0.92 in a forecasting horizon of 30 minutes, representing a significant improvement over baseline models. The study underscores the importance of meaningful pre-processing and model selection in SoC consumption forecasting for AVs, offering insights into future research directions and deployment strategies for enhancing EV efficiency and grid stability.
{"title":"A Comprehensive Leakage-Free Forecasting Pipeline for Segmented Time Series: Application to Cross-Trip State-of-Charge Prediction in Automated Electric Vehicles","authors":"Evangelos Athanasakis;Georgios Spanos;Alexandros Papadopoulos;Antonios Lalas;Konstantinos Votis;Dimitrios Tzovaras","doi":"10.1109/TIV.2024.3519751","DOIUrl":"https://doi.org/10.1109/TIV.2024.3519751","url":null,"abstract":"The rapid adoption of Electric Vehicles (EVs) in the global pursuit of energy efficiency and carbon neutrality necessitates effective strategies to mitigate their carbon footprint and enhance operational stability. Similarly, in order to achieve Sustainability Development Goals, a promising solution toward green mobility, which is gaining ground nowadays, constitutes Automated Vehicles (AVs), which are EVs having the capability to move autonomously, without the need for a driver. One of the most critical factors regarding energy efficiency is the optimal management of energy consumption of AVs. This research study explores the application of machine learning (ML) models for State-of-Charge (SoC) forecasting in AVs, crucial for addressing challenges such as range anxiety and grid overloading. Leveraging real-life EV data from automated minibuses in Gothenburg, Sweeden, a comprehensive pipeline is proposed for data pre-processing, feature selection, and model training. With a focus on predicting SoC several minutes ahead, various ML techniques, including linear regression, ridge regression, lasso regression, and elastic-net regression are embedded in a pipeline specifically developed to overcome the challenge of training time-series models on discontinuous data segments, corresponding to discharge cycles. This pipeline is called Cross-Segment-Leakage-Free (CSLF). The results demonstrate the efficacy of CSLF, with the best-performing model achieving a Mean Absolute Error (MAE) of 0.92 in a forecasting horizon of 30 minutes, representing a significant improvement over baseline models. The study underscores the importance of meaningful pre-processing and model selection in SoC consumption forecasting for AVs, offering insights into future research directions and deployment strategies for enhancing EV efficiency and grid stability.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 12","pages":"5213-5228"},"PeriodicalIF":14.3,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145772098","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}
This paper presents a comprehensive literature review on intelligent driving technologies, with a special emphasis on Automatic Emergency Collision Avoidance Technology (AECA) and Collaborative Stability Control (CSC). These technologies play a crucial role in the active safety of vehicles. AECA proactively detects and responds to potential collisions, and CSC enhances vehicle stability by integrating various systems across multiple driving scenarios. The synergy between AECA and CSC is essential for improving passenger safety and the overall efficiency of traffic systems. This review delves into the application of AECA and CSC, particularly under conditions that might compromise vehicle stability, emphasizing the crucial balance between safety and stability in collision avoidance scenarios. The paper discusses the challenges faced by intelligent vehicles, such as the strong coupling nonlinearity in vehicle dynamics, unpredictable environmental conditions, and the increasing complexity of control systems. It examines strategies in braking, steering, and the coordination of multiple systems to achieve effective collision avoidance and stability control. Additionally, the review provides a forward-looking perspective on potential developments and insights for ongoing research in domains of AECA and CSC within intelligent technologies. The goal is to present a structured overview of the current state of research, highlight significant findings, and identify critical areas where future research could significantly advance the field of intelligent driving systems.
{"title":"Autonomous Emergency Collision Avoidance and Collaborative Stability Control Technologies for Intelligent Vehicles: A Survey","authors":"Xiaoqiang Tan;Guangqiang Wu;Zefan Li;Kai Liu;Chengbao Zhang","doi":"10.1109/TIV.2024.3519766","DOIUrl":"https://doi.org/10.1109/TIV.2024.3519766","url":null,"abstract":"This paper presents a comprehensive literature review on intelligent driving technologies, with a special emphasis on Automatic Emergency Collision Avoidance Technology (AECA) and Collaborative Stability Control (CSC). These technologies play a crucial role in the active safety of vehicles. AECA proactively detects and responds to potential collisions, and CSC enhances vehicle stability by integrating various systems across multiple driving scenarios. The synergy between AECA and CSC is essential for improving passenger safety and the overall efficiency of traffic systems. This review delves into the application of AECA and CSC, particularly under conditions that might compromise vehicle stability, emphasizing the crucial balance between safety and stability in collision avoidance scenarios. The paper discusses the challenges faced by intelligent vehicles, such as the strong coupling nonlinearity in vehicle dynamics, unpredictable environmental conditions, and the increasing complexity of control systems. It examines strategies in braking, steering, and the coordination of multiple systems to achieve effective collision avoidance and stability control. Additionally, the review provides a forward-looking perspective on potential developments and insights for ongoing research in domains of AECA and CSC within intelligent technologies. The goal is to present a structured overview of the current state of research, highlight significant findings, and identify critical areas where future research could significantly advance the field of intelligent driving systems.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 12","pages":"5229-5248"},"PeriodicalIF":14.3,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145772047","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 : 2024-12-17DOI: 10.1109/TIV.2024.3516791
Ji-Ung Im;Jong-Hoon Won
For Advanced Air Mobility (AAM) systems operating in diverse environments, redundant localization techniques are essential to ensure continuous and safe mission execution. In this study, we propose a 3D place recognition and pose estimation method for AAM using a hemispherical light detection and ranging (LiDAR) sensor. The proposed approach includes a feature extraction method that leverages height differences in surrounding objects, a method for generating local and global descriptors from feature distances, and an efficient geometric verification and localization process through correspondence calculation. Additionally, the method incorporates a process to create a virtual descriptor database using a point cloud map, enabling robust localization in unvisited areas. All procedures are handcrafted, and the performance of the proposed method is validated through comparison with state-of-the-art methods using datasets generated in a simulator. The proposed method achieved over 99.16% average precision (AP) and a 99.99% F1 score in loop closure detection. In pose estimation, it achieved a root mean square error (RMSE) of 0.836 meters or less for position and 0.195 degrees or less for heading. Furthermore, a time analysis on both a general PC and an embedded device confirmed the real-time capability of the proposed method, with an average pose estimation time of 21.70 milliseconds on the embedded device, demonstrating its feasibility for real-time localization in low-power environments.
{"title":"Omni Point Air: LiDAR and Point Cloud Map-Based Place Recognition and Pose Estimation for Advanced Air Mobility in GNSS-Denied Environments","authors":"Ji-Ung Im;Jong-Hoon Won","doi":"10.1109/TIV.2024.3516791","DOIUrl":"https://doi.org/10.1109/TIV.2024.3516791","url":null,"abstract":"For Advanced Air Mobility (AAM) systems operating in diverse environments, redundant localization techniques are essential to ensure continuous and safe mission execution. In this study, we propose a 3D place recognition and pose estimation method for AAM using a hemispherical light detection and ranging (LiDAR) sensor. The proposed approach includes a feature extraction method that leverages height differences in surrounding objects, a method for generating local and global descriptors from feature distances, and an efficient geometric verification and localization process through correspondence calculation. Additionally, the method incorporates a process to create a virtual descriptor database using a point cloud map, enabling robust localization in unvisited areas. All procedures are handcrafted, and the performance of the proposed method is validated through comparison with state-of-the-art methods using datasets generated in a simulator. The proposed method achieved over 99.16% average precision (AP) and a 99.99% F1 score in loop closure detection. In pose estimation, it achieved a root mean square error (RMSE) of 0.836 meters or less for position and 0.195 degrees or less for heading. Furthermore, a time analysis on both a general PC and an embedded device confirmed the real-time capability of the proposed method, with an average pose estimation time of 21.70 milliseconds on the embedded device, demonstrating its feasibility for real-time localization in low-power environments.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 12","pages":"5162-5176"},"PeriodicalIF":14.3,"publicationDate":"2024-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145772052","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 : 2024-12-16DOI: 10.1109/TIV.2024.3517880
Daniel Louback S. Lubanco;Ahmed Hashem;Markus Pichler-Scheder;Thomas Schlechter;Reinhard Feger;Andreas Stelzer
In this paper we propose a radar-only simultaneous localization and mapping algorithm based on multiple input multiple output synthetic aperture radar images. The algorithm distinguishes itself from others by depending only on radar data for generating synthetic aperture radar images for estimating traversed trajectory and building a visual representation. In our algorithm, ego-velocity (estimated using only radar data) is used for generating synthetic aperture radar images. The generated radar images are used for rotation estimation in the odometry step as well as for place recognition by exploiting the Fourier-Radon image registration approach. After the trajectory is optimized, we combine coherent and incoherent processing over the radar data for generating a map of the traversed area. The proposed concept was evaluated over multiple sequences comprising heterogeneous and dynamic environments. The results show high performance of the algorithm in terms of place recognition, attaining a balanced f-score in the range of 0.86–0.96. Moreover, the algorithm also achieves good results in terms of simultaneous localization and mapping. For example, it achieves an absolute trajectory error of 0.11 m for a trajectory of length 340 m, and 0.43 m for a trajectory of length 1092 m. Finally, we also include a case study in which we show the capability of the radar-only localization and mapping solution in operating under scenarios that are challenging for global navigation satellite systems.
{"title":"MS-SLAM: Multiple Input Multiple Output Synthetic Aperture Radar Simultaneous Localization and Mapping","authors":"Daniel Louback S. Lubanco;Ahmed Hashem;Markus Pichler-Scheder;Thomas Schlechter;Reinhard Feger;Andreas Stelzer","doi":"10.1109/TIV.2024.3517880","DOIUrl":"https://doi.org/10.1109/TIV.2024.3517880","url":null,"abstract":"In this paper we propose a radar-only simultaneous localization and mapping algorithm based on multiple input multiple output synthetic aperture radar images. The algorithm distinguishes itself from others by depending only on radar data for generating synthetic aperture radar images for estimating traversed trajectory and building a visual representation. In our algorithm, ego-velocity (estimated using only radar data) is used for generating synthetic aperture radar images. The generated radar images are used for rotation estimation in the odometry step as well as for place recognition by exploiting the Fourier-Radon image registration approach. After the trajectory is optimized, we combine coherent and incoherent processing over the radar data for generating a map of the traversed area. The proposed concept was evaluated over multiple sequences comprising heterogeneous and dynamic environments. The results show high performance of the algorithm in terms of place recognition, attaining a balanced f-score in the range of 0.86–0.96. Moreover, the algorithm also achieves good results in terms of simultaneous localization and mapping. For example, it achieves an absolute trajectory error of 0.11 m for a trajectory of length 340 m, and 0.43 m for a trajectory of length 1092 m. Finally, we also include a case study in which we show the capability of the radar-only localization and mapping solution in operating under scenarios that are challenging for global navigation satellite systems.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 12","pages":"5177-5196"},"PeriodicalIF":14.3,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10803095","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145772022","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-13DOI: 10.1109/TIV.2024.3506727
Chi Zhang;Janis Sprenger;Zhongjun Ni;Christian Berger
Predicting pedestrian crossing behavior is important for intelligent traffic systems to avoid pedestrian-vehicle collisions. Most existing pedestrian crossing behavior models are trained and evaluated on datasets collected from a single country, overlooking differences between countries. To address this gap, we compared pedestrian road-crossing behavior at unsignalized crossings in Germany and Japan. We presented four types of machine learning models to predict gap selection behavior, zebra crossing usage, and their trajectories using simulator data collected from both countries. When comparing the differences between countries, pedestrians from the study conducted in Japan are more cautious, selecting larger gaps compared to those in Germany. We evaluate and analyze model transferability. Our results show that neural networks outperform other machine learning models in predicting gap selection and zebra crossing usage, while random forest models perform best on trajectory prediction tasks, demonstrating strong performance and transferability. We develop a transferable model using an unsupervised clustering method, which improves prediction accuracy for gap selection and trajectory prediction. These findings provide a deeper understanding of pedestrian crossing behaviors in different countries and offer valuable insights into model transferability.
{"title":"Predicting Pedestrian Crossing Behavior in Germany and Japan: Insights Into Model Transferability","authors":"Chi Zhang;Janis Sprenger;Zhongjun Ni;Christian Berger","doi":"10.1109/TIV.2024.3506727","DOIUrl":"https://doi.org/10.1109/TIV.2024.3506727","url":null,"abstract":"Predicting pedestrian crossing behavior is important for intelligent traffic systems to avoid pedestrian-vehicle collisions. Most existing pedestrian crossing behavior models are trained and evaluated on datasets collected from a single country, overlooking differences between countries. To address this gap, we compared pedestrian road-crossing behavior at unsignalized crossings in Germany and Japan. We presented four types of machine learning models to predict gap selection behavior, zebra crossing usage, and their trajectories using simulator data collected from both countries. When comparing the differences between countries, pedestrians from the study conducted in Japan are more cautious, selecting larger gaps compared to those in Germany. We evaluate and analyze model transferability. Our results show that neural networks outperform other machine learning models in predicting gap selection and zebra crossing usage, while random forest models perform best on trajectory prediction tasks, demonstrating strong performance and transferability. We develop a transferable model using an unsupervised clustering method, which improves prediction accuracy for gap selection and trajectory prediction. These findings provide a deeper understanding of pedestrian crossing behaviors in different countries and offer valuable insights into model transferability.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 11","pages":"4887-4902"},"PeriodicalIF":14.3,"publicationDate":"2024-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10798977","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145665806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","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：