Pub Date : 2024-09-18DOI: 10.1109/tits.2024.3453769
Lin Yu, Lei Qiao, Chao Shen
{"title":"Obstacle Avoidance for a Large-Scale High-Speed Underactuated AUV in Complex Environments","authors":"Lin Yu, Lei Qiao, Chao Shen","doi":"10.1109/tits.2024.3453769","DOIUrl":"https://doi.org/10.1109/tits.2024.3453769","url":null,"abstract":"","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"6 1","pages":""},"PeriodicalIF":8.5,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264265","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 : 2024-09-18DOI: 10.1109/TITS.2024.3421917
Zixian Zhao;Xingchen Zhang;Yiannis Demiris
In the era of artificial intelligence, privacy has become a paramount concern, especially within intelligent transportation systems (ITS) where pedestrians are frequently captured by vehicle-mounted cameras for deep learning model training. To address this, we introduce 3PFS, a novel method designed to protect pedestrian privacy via face swapping while preserving the utility of processed images. Our method consists of a pedestrian detector, a face detector, a pre-processing module, a source face selection algorithm, and a face swapping algorithm. After detecting pedestrians and their corresponding faces, the pre-processing module enhances image quality. Our unique source face selection algorithm then chooses an appropriate face from our source face library, which is subsequently swapped with the target face using a face swapping algorithm. Notably, with the combination of a pedestrian tracking algorithm, our 3PFS is well-suited for video anonymization. Additionally, we propose a comprehensive evaluation strategy to evaluate the performance of pedestrian anonymization methods. We validate the effectiveness of 3PFS through extensive experiments on a dataset we created based on the publicly available JAAD dataset and on videos captured using our robotic wheelchair.
{"title":"3PFS: Protecting Pedestrian Privacy Through Face Swapping","authors":"Zixian Zhao;Xingchen Zhang;Yiannis Demiris","doi":"10.1109/TITS.2024.3421917","DOIUrl":"10.1109/TITS.2024.3421917","url":null,"abstract":"In the era of artificial intelligence, privacy has become a paramount concern, especially within intelligent transportation systems (ITS) where pedestrians are frequently captured by vehicle-mounted cameras for deep learning model training. To address this, we introduce 3PFS, a novel method designed to protect pedestrian privacy via face swapping while preserving the utility of processed images. Our method consists of a pedestrian detector, a face detector, a pre-processing module, a source face selection algorithm, and a face swapping algorithm. After detecting pedestrians and their corresponding faces, the pre-processing module enhances image quality. Our unique source face selection algorithm then chooses an appropriate face from our source face library, which is subsequently swapped with the target face using a face swapping algorithm. Notably, with the combination of a pedestrian tracking algorithm, our 3PFS is well-suited for video anonymization. Additionally, we propose a comprehensive evaluation strategy to evaluate the performance of pedestrian anonymization methods. We validate the effectiveness of 3PFS through extensive experiments on a dataset we created based on the publicly available JAAD dataset and on videos captured using our robotic wheelchair.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"16845-16854"},"PeriodicalIF":7.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264261","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 : 2024-09-18DOI: 10.1109/TITS.2024.3443341
Yiling Wu;Yingping Zhao;Xinfeng Zhang;Yaowei Wang
Traffic demand prediction has been drawing increasing research interest due to its critical role in intelligent transportation systems. However, conventional deep learning methods for traffic demand forecast ignore the correlations between the pick-up and drop-off demands, thus not fully exploring the patterns of demand evolution. In this work, the pick-up and drop-off demands are treated as two modalities, and an architecture is designed to explicitly model the interactions between the pick-up and drop-off demands both spatially and temporally. Specifically, the self-attention mechanism is adopted to automatically discover spatio-temporal patterns without manual designation for each demand. Then, the cross-attention mechanism is utilized to let the two demands attend to each other, resulting in information exchange between the two demands. The self-attention and cross-attention are combined to capture spatio-temporal correlations simultaneously. Finally, experiments are carried out on three real-world datasets, NYC Citi Bike, NYC Taxi, and BJ Subway, and the results show that this newly proposed method outperforms the state-of-the-art methods.
{"title":"Spatial-Temporal Correlation Learning for Traffic Demand Prediction","authors":"Yiling Wu;Yingping Zhao;Xinfeng Zhang;Yaowei Wang","doi":"10.1109/TITS.2024.3443341","DOIUrl":"10.1109/TITS.2024.3443341","url":null,"abstract":"Traffic demand prediction has been drawing increasing research interest due to its critical role in intelligent transportation systems. However, conventional deep learning methods for traffic demand forecast ignore the correlations between the pick-up and drop-off demands, thus not fully exploring the patterns of demand evolution. In this work, the pick-up and drop-off demands are treated as two modalities, and an architecture is designed to explicitly model the interactions between the pick-up and drop-off demands both spatially and temporally. Specifically, the self-attention mechanism is adopted to automatically discover spatio-temporal patterns without manual designation for each demand. Then, the cross-attention mechanism is utilized to let the two demands attend to each other, resulting in information exchange between the two demands. The self-attention and cross-attention are combined to capture spatio-temporal correlations simultaneously. Finally, experiments are carried out on three real-world datasets, NYC Citi Bike, NYC Taxi, and BJ Subway, and the results show that this newly proposed method outperforms the state-of-the-art methods.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"15745-15758"},"PeriodicalIF":7.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264262","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 : 2024-09-18DOI: 10.1109/TITS.2024.3452647
Ming Chen;Ratnasingham Tharmarasa;Thia Kirubarajan;Sunil Chomal
Autonomous vehicle technology is rapidly becoming the driving force in the automobile industry. As such, the interest in high-resolution radio detection and ranging (radar) for autonomous vehicle applications is increasing due to its affordability and high angular resolution. However, for Advanced Driver Assistance Systems (ADAS), the challenge of azimuth ambiguity caused by a large physical distance between radar antennas is prevalent. This causes false measurements in a direction different from the target’s true angle due to grating lobes. This challenge increases when extended targets are considered. This paper proposes a Pseudo-3D Assignment (P3DA) method based on a Pseudo Measurement Set (PMS) to resolve azimuth ambiguity in multiple extended target tracking. The proposed method can resolve mono (single) and split (duplicated) azimuth ambiguities common in extended target tracking. The proposed solution uses Lagrangian Relaxation based on a Flexible Search (LR-FS) algorithm to solve the P3DA-PMS problem efficiently. The performance of the proposed algorithm in a typical traffic scenario simulated in Unreal Engine 4, with an ego vehicle mounted with both 2D (unambiguous) and 3D (ambiguous) radars, is evaluated. Simulation and experiment results suggest that the proposed P3DA-PMS-based tracking algorithm can perform better than conventional methods.
自动驾驶汽车技术正迅速成为汽车行业的驱动力。因此，高分辨率无线电探测和测距（雷达）因其经济实惠和高角度分辨率而越来越受到自动驾驶汽车应用的关注。然而，对于高级驾驶辅助系统（ADAS）来说，雷达天线之间的物理距离过大造成的方位模糊是一个普遍存在的挑战。由于光栅裂片的影响，这会造成与目标真实角度不同方向的错误测量。当考虑到扩展目标时，这一挑战就更大了。本文提出了一种基于伪测量集（PMS）的伪三维赋值（P3DA）方法，以解决多扩展目标跟踪中的方位角模糊问题。所提出的方法可以解决扩展目标跟踪中常见的单（单一）和分（重复）方位角模糊问题。提出的解决方案采用基于灵活搜索（LR-FS）的拉格朗日松弛算法来高效解决 P3DA-PMS 问题。在虚幻引擎 4 中模拟的一个典型交通场景中，评估了所提算法的性能，该场景中的自我车辆同时安装了 2D（不明确）和 3D （模糊）雷达。仿真和实验结果表明，所提出的基于 P3DA-PMS 的跟踪算法比传统方法性能更好。
{"title":"An Assignment Method for Multiple Extended Target Tracking With Azimuth Ambiguity Based on Pseudo Measurement Set","authors":"Ming Chen;Ratnasingham Tharmarasa;Thia Kirubarajan;Sunil Chomal","doi":"10.1109/TITS.2024.3452647","DOIUrl":"10.1109/TITS.2024.3452647","url":null,"abstract":"Autonomous vehicle technology is rapidly becoming the driving force in the automobile industry. As such, the interest in high-resolution radio detection and ranging (radar) for autonomous vehicle applications is increasing due to its affordability and high angular resolution. However, for Advanced Driver Assistance Systems (ADAS), the challenge of azimuth ambiguity caused by a large physical distance between radar antennas is prevalent. This causes false measurements in a direction different from the target’s true angle due to grating lobes. This challenge increases when extended targets are considered. This paper proposes a Pseudo-3D Assignment (P3DA) method based on a Pseudo Measurement Set (PMS) to resolve azimuth ambiguity in multiple extended target tracking. The proposed method can resolve mono (single) and split (duplicated) azimuth ambiguities common in extended target tracking. The proposed solution uses Lagrangian Relaxation based on a Flexible Search (LR-FS) algorithm to solve the P3DA-PMS problem efficiently. The performance of the proposed algorithm in a typical traffic scenario simulated in Unreal Engine 4, with an ego vehicle mounted with both 2D (unambiguous) and 3D (ambiguous) radars, is evaluated. Simulation and experiment results suggest that the proposed P3DA-PMS-based tracking algorithm can perform better than conventional methods.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"15512-15531"},"PeriodicalIF":7.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264263","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 development of intelligent transportation systems (ITS) has attracted significant attention to connected and autonomous vehicles (CAVs). It is urgent to investigate multi-CAV intelligent cruise control solutions in mixed traffic environments. In addition, the impact of platoon dynamics and time delays, induced by shared wireless communications, data processing, and actuation cannot be ignored. This article investigates the development of a multi-agent deep reinforcement learning (MADRL) controller tailored for CAVs operating within mixed and dynamic traffic scenarios that involve time delays. Firstly, the error dynamics in the discrete-time domain for each subplatoon is derived by considering the time-varying delays and leading vehicle states, and then the optimal CAV cruise control problem is formulated. Subsequently, the partially observable Markov game (POMG) is used to construct the multi-agent environment, and then a centralized training decentralized execution (CTDE) algorithm framework is proposed based on the multi-agent deep deterministic policy gradient (MADDPG) method. Finally, the computational complexity and the influence of delay are analyzed. The simulation results illustrate the effectiveness of the proposed intelligent algorithm.
{"title":"Multi-Agent DRL-Controlled Connected and Automated Vehicles in Mixed Traffic With Time Delays","authors":"Zhuwei Wang;Yi Xue;Lihan Liu;Haijun Zhang;Chunhui Qu;Chao Fang","doi":"10.1109/TITS.2024.3435036","DOIUrl":"10.1109/TITS.2024.3435036","url":null,"abstract":"The development of intelligent transportation systems (ITS) has attracted significant attention to connected and autonomous vehicles (CAVs). It is urgent to investigate multi-CAV intelligent cruise control solutions in mixed traffic environments. In addition, the impact of platoon dynamics and time delays, induced by shared wireless communications, data processing, and actuation cannot be ignored. This article investigates the development of a multi-agent deep reinforcement learning (MADRL) controller tailored for CAVs operating within mixed and dynamic traffic scenarios that involve time delays. Firstly, the error dynamics in the discrete-time domain for each subplatoon is derived by considering the time-varying delays and leading vehicle states, and then the optimal CAV cruise control problem is formulated. Subsequently, the partially observable Markov game (POMG) is used to construct the multi-agent environment, and then a centralized training decentralized execution (CTDE) algorithm framework is proposed based on the multi-agent deep deterministic policy gradient (MADDPG) method. Finally, the computational complexity and the influence of delay are analyzed. The simulation results illustrate the effectiveness of the proposed intelligent algorithm.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"17676-17688"},"PeriodicalIF":7.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264175","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}
Traffic flow forecasting is challenging due to the intricate spatio-temporal correlations in traffic patterns. Previous works captured spatial dependencies based on graph neural networks and used fixed graph construction methods to characterize spatial relationships, which limits the ability of models to capture dynamic and long-range spatial dependencies. Meanwhile, prior studies did not consider the issue of a large number of redundant parameters in traffic prediction models, which not only increases the storage cost of the model but also reduces its generalization ability. To address the above challenges, we propose a Dynamic Spatial Transformer for Traffic Forecasting with Low-Rank Tensor Compression (DSFormer-LRTC). Specifically, we constructed a global spatial Transformer to capture remote spatial dependencies, and a distance-based mask matrix is used in local spatial Transformer to enhance the adjacent spatial influence. To reduce the complexity of the model, the model adopts a design that separates temporal and spatial. Meanwhile, we introduce low-rank tensor decomposition to reconstruct the parameter matrix in Transformer module to compress the proposed model. Experimental results show that DSFormer-LRTC achieves state-of-the-art performance on four real-world datasets. The experimental analysis of attention matrix also proves that the model can learn dynamic and distant spatial features. Finally, the compressed model parameters reduce the original parameter size by two-thirds, while significantly outperforming the baseline model in terms of computational efficiency.
{"title":"DSFormer-LRTC: Dynamic Spatial Transformer for Traffic Forecasting With Low-Rank Tensor Compression","authors":"Jianli Zhao;Futong Zhuo;Qiuxia Sun;Qing Li;Yiran Hua;Jianye Zhao","doi":"10.1109/TITS.2024.3436523","DOIUrl":"10.1109/TITS.2024.3436523","url":null,"abstract":"Traffic flow forecasting is challenging due to the intricate spatio-temporal correlations in traffic patterns. Previous works captured spatial dependencies based on graph neural networks and used fixed graph construction methods to characterize spatial relationships, which limits the ability of models to capture dynamic and long-range spatial dependencies. Meanwhile, prior studies did not consider the issue of a large number of redundant parameters in traffic prediction models, which not only increases the storage cost of the model but also reduces its generalization ability. To address the above challenges, we propose a Dynamic Spatial Transformer for Traffic Forecasting with Low-Rank Tensor Compression (DSFormer-LRTC). Specifically, we constructed a global spatial Transformer to capture remote spatial dependencies, and a distance-based mask matrix is used in local spatial Transformer to enhance the adjacent spatial influence. To reduce the complexity of the model, the model adopts a design that separates temporal and spatial. Meanwhile, we introduce low-rank tensor decomposition to reconstruct the parameter matrix in Transformer module to compress the proposed model. Experimental results show that DSFormer-LRTC achieves state-of-the-art performance on four real-world datasets. The experimental analysis of attention matrix also proves that the model can learn dynamic and distant spatial features. Finally, the compressed model parameters reduce the original parameter size by two-thirds, while significantly outperforming the baseline model in terms of computational efficiency.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"16323-16335"},"PeriodicalIF":7.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264172","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 : 2024-09-18DOI: 10.1109/tits.2024.3447357
Mubashir Murshed, Glaucio H. S. Carvalho, Robson E. De Grande
{"title":"Ensemble SARSA and LSTM for User-Centric Handover Decisions in 5G Vehicular Networks","authors":"Mubashir Murshed, Glaucio H. S. Carvalho, Robson E. De Grande","doi":"10.1109/tits.2024.3447357","DOIUrl":"https://doi.org/10.1109/tits.2024.3447357","url":null,"abstract":"","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"50 1","pages":""},"PeriodicalIF":8.5,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264267","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 : 2024-09-18DOI: 10.1109/tits.2024.3449336
Miao Yu, Hao Zhang, Jiaqi Li, Shuling Kang, Mingyuan Gao, Jin Zhu, Jian Fu
{"title":"The Application of 6G Technology and Back Propagation Neural Network in the Smart Public Transport System","authors":"Miao Yu, Hao Zhang, Jiaqi Li, Shuling Kang, Mingyuan Gao, Jin Zhu, Jian Fu","doi":"10.1109/tits.2024.3449336","DOIUrl":"https://doi.org/10.1109/tits.2024.3449336","url":null,"abstract":"","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"16 1","pages":""},"PeriodicalIF":8.5,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264268","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}
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：