IEEE Transactions on Intelligent Vehicles最新文献_第8页

The Integration of Prediction and Planning in Deep Learning Automated Driving Systems: A Review 深度学习自动驾驶系统中预测与规划的集成研究综述

IF 14.3 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-11 DOI: 10.1109/TIV.2024.3459071

Steffen Hagedorn;Marcel Hallgarten;Martin Stoll;Alexandru Paul Condurache

{"title":"The Integration of Prediction and Planning in Deep Learning Automated Driving Systems: A Review","authors":"Steffen Hagedorn;Marcel Hallgarten;Martin Stoll;Alexandru Paul Condurache","doi":"10.1109/TIV.2024.3459071","DOIUrl":"https://doi.org/10.1109/TIV.2024.3459071","url":null,"abstract":"Automated driving has the potential to revolutionize personal, public, and freight mobility. Beside accurately perceiving the environment, automated vehicles must plan a safe, comfortable, and efficient motion trajectory. To promote safety and progress, many works rely on modules that predict the future motion of surrounding traffic. Modular automated driving systems commonly handle prediction and planning as sequential, separate tasks. While this accounts for the influence of surrounding traffic on the ego vehicle, it fails to anticipate the reactions of traffic participants to the ego vehicle's behavior. Recent methods increasingly integrate prediction and planning in a joint or interdependent step to model bidirectional interactions. To date, a comprehensive overview of different integration principles is lacking. We systematically review state-of-the-art deep learning-based planning systems, and focus on how they integrate prediction. Different facets of the integration ranging from system architecture to high-level behavioral aspects are considered and related to each other. Moreover, we discuss the implications, strengths, and limitations of different integration principles. By pointing out research gaps, describing relevant future challenges, and highlighting trends in the research field, we identify promising directions for future research.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 5","pages":"3626-3643"},"PeriodicalIF":14.3,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144990035","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}

引用次数: 0

Online, Target-Free LiDAR-Camera Extrinsic Calibration via Cross-Modal Mask Matching 基于跨模态掩模匹配的在线无目标激光雷达相机外部标定

IF 14.3 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-11 DOI: 10.1109/TIV.2024.3456299

Zhiwei Huang;Yikang Zhang;Qijun Chen;Rui Fan

{"title":"Online, Target-Free LiDAR-Camera Extrinsic Calibration via Cross-Modal Mask Matching","authors":"Zhiwei Huang;Yikang Zhang;Qijun Chen;Rui Fan","doi":"10.1109/TIV.2024.3456299","DOIUrl":"https://doi.org/10.1109/TIV.2024.3456299","url":null,"abstract":"LiDAR-camera extrinsic calibration (LCEC) is crucial for data fusion in intelligent vehicles. Offline, target-based approaches have long been the preferred choice in this field. However, they often demonstrate poor adaptability to real-world environments. This is largely because extrinsic parameters may change significantly due to moderate shocks or during extended operations in environments with vibrations. In contrast, online, target-free approaches provide greater adaptability yet typically lack robustness, primarily due to the challenges in cross-modal feature matching. Therefore, in this article, we unleash the full potential of large vision models (LVMs), which are emerging as a significant trend in the fields of computer vision and robotics, especially for embodied artificial intelligence, to achieve robust and accurate online, target-free LCEC across a variety of challenging scenarios. Our main contributions are threefold: we introduce a novel framework known as MIAS-LCEC, provide an open-source versatile calibration toolbox with an interactive visualization interface, and publish three real-world datasets captured from various indoor and outdoor environments. The cornerstone of our framework and toolbox is the cross-modal mask matching (C3M) algorithm, developed based on a state-of-the-art (SoTA) LVM and capable of generating sufficient and reliable matches. Extensive experiments conducted on these real-world datasets demonstrate the robustness of our approach and its superior performance compared to SoTA methods, particularly for the solid-state LiDARs with super-wide fields of view. Our toolbox and datasets are publicly available at <uri>https://mias.group/MIAS-LCEC</uri>.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 5","pages":"3531-3542"},"PeriodicalIF":14.3,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144990078","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}

引用次数: 0

Investigation of RBF-SMC Control Strategy for Vertical Dynamics of Maglev Car Considering Temperature Rise Effects 考虑温升效应的磁悬浮车垂直动力学RBF-SMC控制策略研究

IF 14.3 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-10 DOI: 10.1109/TIV.2024.3456784

Zhengyan Li;Zhihao Ke;Jiaheng Shi;Hongfu Shi;Zigang Deng

{"title":"Investigation of RBF-SMC Control Strategy for Vertical Dynamics of Maglev Car Considering Temperature Rise Effects","authors":"Zhengyan Li;Zhihao Ke;Jiaheng Shi;Hongfu Shi;Zigang Deng","doi":"10.1109/TIV.2024.3456784","DOIUrl":"https://doi.org/10.1109/TIV.2024.3456784","url":null,"abstract":"To address the levitation force attenuation in a magnetic levitation (maglev) car with a permanent magnet electrodynamic wheel (PMEDW) system caused by the temperature rise of the conductive plate, this paper proposes an adaptive sliding mode control strategy combined with a Radial Basis Function neural network (RBF-SMC). This approach enhances both the levitation stability and anti-interference capability of the maglev car system. Initially, a four-wheel dynamic model is established. The RBF neural network is then introduced to observe and mitigate disturbances caused by the temperature rise. An RBF-SMC control strategy is designed to improve the system's static levitation stability. The effectiveness of this control strategy is evaluated through simulations and experiments under various operating conditions. The research results indicate that, compared to the traditional PID control strategy, the proposed method reduces tracking error by 75.2%, compensates for the levitation force attenuation caused by the eddy current temperature rise of the conductive plate, and suppresses disturbances.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 5","pages":"3560-3572"},"PeriodicalIF":14.3,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144990207","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}

引用次数: 0

An Online Energy Management System Based on Minimum-Time Speed Planning for Autonomous Underwater Vehicles 基于最小时间速度规划的自主水下航行器在线能量管理系统

IF 14.3 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-10 DOI: 10.1109/TIV.2024.3457688

Seyed Nima Hosseini Eimeni;Alireza Khosravi

{"title":"An Online Energy Management System Based on Minimum-Time Speed Planning for Autonomous Underwater Vehicles","authors":"Seyed Nima Hosseini Eimeni;Alireza Khosravi","doi":"10.1109/TIV.2024.3457688","DOIUrl":"https://doi.org/10.1109/TIV.2024.3457688","url":null,"abstract":"Velocity of autonomous underwater vehicles (AUVs) plays a significant role in the energy consumption of these vehicles, as well as their other capabilities, such as localization, maneuverability, and time to complete assigned missions. In this paper, for the first time, an online energy management strategy (EMS) is proposed for AUVs, which enables approaching the target in minimum time by adjusting the sub-optimal speed according to the amount of available energy. In this model-based energy management system, a framework is proposed that its output is the highest forward speed, so that the mission duration is minimized and the stored battery energy is used in the best possible way. In this field, existing efforts are mostly focused on reducing energy consumption, without considering other vehicle capabilities, but this method tries to reduce this gap. First, the architecture of an AUV propulsion is described and modeled, and a nonlinear equation is derived to generate the sub-optimal speed, which simultaneously minimize the amount of energy consumption and the time to reach the target. Then, for online implementation of this algorithm, a framework is proposed that generates the desired speed by using the battery state of charge (SOC), the voltage and the instantaneous Distance to destination. Illustrative simulation examples were conducted in MATLAB/Simulink to demonstrate the validity of the proposed scheme and the hardware in the loop test is conducted to evaluate the computational complexity of algorithm. Finally, experimental results showed the practical effectiveness of the proposed EMS.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 5","pages":"3600-3612"},"PeriodicalIF":14.3,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144990050","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}

引用次数: 0

Adaptive Square-Root Cubature Kalman Filter Based Low Cost UAV Positioning in Dark and GPS-Denied Environments 基于自适应平方根培养卡尔曼滤波的无人机在黑暗和gps拒绝环境下的定位

IF 14.3 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-10 DOI: 10.1109/TIV.2024.3457678

Beiya Yang;Erfu Yang;Haobin Shi;Leijian Yu;Cong Niu

{"title":"Adaptive Square-Root Cubature Kalman Filter Based Low Cost UAV Positioning in Dark and GPS-Denied Environments","authors":"Beiya Yang;Erfu Yang;Haobin Shi;Leijian Yu;Cong Niu","doi":"10.1109/TIV.2024.3457678","DOIUrl":"https://doi.org/10.1109/TIV.2024.3457678","url":null,"abstract":"Routine inspection inside the water tank, pressure vessel, penstocks and boiler which present dark and global positioning system (GPS) denied environment always plays an important role for the safety storage and transportation. The conventional inspection conducted by the skilled workers is highly expensive, time consuming and may cause the safety and heath problem. Nowadays, the emerging unmanned aerial vehicle (UAV) based techniques make it possible to replace human to do the periodical inspection in these environments. However, how to obtain the reliable, high accuracy and precise position information of the UAV becomes a challenging issue, as the GPS is unable to provide the accurate position information in these environments. In order to resolve this problem, an adaptive square-root cubature Kalman filter (ASRCKF) based low cost UAV positioning system is designed. Through the combination of the inertial measurement unit (IMU), ultra-wideband (UWB), the cubature rule, the adaptively estimated noise model and weighting factors, the potential degradation and oscillation for the system performance which caused by the linearisation process, the variation of the measurement noise and the manually adjusted noise model are solved. Finally, the 0.081m median localisation error, 0.172m 95<inline-formula><tex-math>$^{th}$</tex-math></inline-formula> percentile localisation error and 0.045m average standard deviation (STD) can be attained, which can support the UAV to achieve the autonomous inspection in dark and GPS-denied environments.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 5","pages":"3587-3599"},"PeriodicalIF":14.3,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144990195","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}

引用次数: 0

Smart Mobility With Agent-Based Foundation Models: Towards Interactive and Collaborative Intelligent Vehicles 基于代理基础模型的智能交通：实现交互式协作智能汽车

IF 14 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-10 DOI: 10.1109/TIV.2024.3457759

Bingyi Xia;Peijia Xie;Jiankun Wang

{"title":"Smart Mobility With Agent-Based Foundation Models: Towards Interactive and Collaborative Intelligent Vehicles","authors":"Bingyi Xia;Peijia Xie;Jiankun Wang","doi":"10.1109/TIV.2024.3457759","DOIUrl":"https://doi.org/10.1109/TIV.2024.3457759","url":null,"abstract":"This letter reports the insights gained during a Distributed/Decentralized Hybrid Workshop on Foundation/Infrastructure Intelligence (FII), where we discussed the evolving role of Foundation Models in the field of intelligent vehicles. These models, pre-trained on multimodal data, have emerged as pivotal in the landscape of intelligent vehicles by leveraging their capabilities for high-level reasoning. Ongoing research focuses on these models to further improve scene perception and decision-making, aiming to develop adaptive systems for robot navigation and autonomous driving. However, for smart mobility across the Cyber-Physical-Social space, foundation intelligence should learn human-level knowledge to perform sophisticated interactions and collaborations based on human feedback. Agent-based Foundation Models, as the new training paradigm, can generate cross-domain actions consistent with perception information, paving the way to realize interactive and collaborative agents. This letter discusses the challenges of enhancing and leveraging the scene understanding and spatial reasoning capabilities of the pre-trained foundation model for smart mobility. It also offers insights into the embodied employment of foundation and infrastructure intelligence in enhancing multimodal interactions between robots, environments, and humans.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"9 7","pages":"5130-5133"},"PeriodicalIF":14.0,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142320503","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}

引用次数: 0

Adaptive Depth Graph Neural Network-Based Dynamic Task Allocation for UAV-UGVs Under Complex Environments 基于自适应深度图神经网络的复杂环境下无人机- ugv动态任务分配

IF 14.3 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-10 DOI: 10.1109/TIV.2024.3457493

Ziyuan Ma;Jun Xiong;Huajun Gong;Xinhua Wang

{"title":"Adaptive Depth Graph Neural Network-Based Dynamic Task Allocation for UAV-UGVs Under Complex Environments","authors":"Ziyuan Ma;Jun Xiong;Huajun Gong;Xinhua Wang","doi":"10.1109/TIV.2024.3457493","DOIUrl":"https://doi.org/10.1109/TIV.2024.3457493","url":null,"abstract":"This paper investigates dynamic task allocation (DTA) for unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) in complex urban environments using an adaptive depth graph neural network (AD-GNN) combined with biomimetic algorithms. The goal is to improve collaborative operational efficiency for UAVs and UGVs engaged in reconnaissance, combat, relay, suicide, and electronic warfare tasks. The proposed approach validates through scenarios like urban search and rescue, counterterrorism, and disaster management. AD-GNN dynamically adjusts its depth based on scenario complexity, while biomimetic algorithms optimize task allocation by emulating natural processes. Existing methods often focus on static task allocation or separate enhancement of UAV and UGV capabilities, lacking adaptability and efficiency in unpredictable urban settings. The AD-GNN model addresses these limitations by adjusting its complexity in real-time, optimizing decision-making for task allocation. The integration of biomimetic algorithms enhances robustness and flexibility, adapting to changing operational demands. Simulations using real-world geographic information system (GIS) data in extensive urban settings demonstrate significant improvements in task allocation efficiency. UAVs and UGVs achieve operational efficiencies above 85% in search and rescue operations and 90–95% in disaster management scenarios after optimization. These results highlight the model's capability to manage and allocate tasks efficiently in dynamic and unpredictable urban environments. In conclusion, this paper contributes to autonomous systems by offering an innovative solution for DTA in urban settings, showcasing the potential of integrating advanced graph neural networks with biomimetic principles to enhance UAV and UGV fleet operations in complex environments.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 5","pages":"3573-3586"},"PeriodicalIF":14.3,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144990045","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}

引用次数: 0

Pedestrian-Vehicle Information Modulation for Pedestrian Crossing Intention Prediction 行人过马路意图预测的行人-车辆信息调制

IF 14.3 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-09 DOI: 10.1109/TIV.2024.3437779

Li Xu;Shaodi You;Gang He;Yunsong Li

引用次数: 0

MGSGNet-S*: Multilayer Guided Semantic Graph Network via Knowledge Distillation for RGB-Thermal Urban Scene Parsing MGSGNet-S*：基于知识蒸馏的多层引导语义图网络rgb -热城市场景分析

IF 14.3 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-09 DOI: 10.1109/TIV.2024.3456437

Wujie Zhou;Hongping Wu;Qiuping Jiang

{"title":"MGSGNet-S*: Multilayer Guided Semantic Graph Network via Knowledge Distillation for RGB-Thermal Urban Scene Parsing","authors":"Wujie Zhou;Hongping Wu;Qiuping Jiang","doi":"10.1109/TIV.2024.3456437","DOIUrl":"https://doi.org/10.1109/TIV.2024.3456437","url":null,"abstract":"Owing to rapid developments in driverless technologies, vision tasks for unmanned vehicles have gained considerable attention, particularly in multimodal-based urban scene parsing. Although deep-learning algorithms have outperformed traditional models in such tasks, they cannot operate on mobile devices and edge networks owing to the coarse-grained cross-modal complementary information alignment, inadequate modeling of semantic-category relations, overabundance of parameters, and high computational complexity. To address these issues, a multilayer guided semantic graph network via knowledge distillation (MGSGNet-S<sup>*</sup>) is proposed for red-green-blue-thermal urban scene parsing. First, a new cross-modal adaptive fusion module adjusts pixel-level adaptive modal complementary information by incorporating additional deep modal information and residual cross-modal matrix fine-grained attention. Second, a novel semantic graph module overcomes the misclassification problems of objects of the same semantic class during low-level encoding by incorporating high-level information in the Euclidean space and modeling semantic graph relationships in the non-Euclidean space. Finally, to strike the balance between accuracy and efficiency, a tailored framework optimally utilizes effective knowledge of pixel intra- and inter-class similarity, fusion features, and cross-modal correlation. Experimental results indicate that MGSGNet-S<sup>*</sup> considerably outperforms relevant state-of-the-art methods with fewer parameters and lower computational costs. The numbers of parameters and floating-point operations were reduced by 95.69% and 93.34%, respectively, relative to those for the teacher model, thus demonstrating stronger inferencing capabilities at 28.65 frames per second.","PeriodicalId":36532,"journal":{"name":"IEEE Transactions on Intelligent Vehicles","volume":"10 5","pages":"3543-3559"},"PeriodicalIF":14.3,"publicationDate":"2024-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144990146","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}

引用次数: 0

Learning Yaw Velocity for Inertial-Wheel Odometry on Autonomous Vehicles 自动驾驶汽车惯性车轮里程测量的偏航速度学习

IF 14.3 1区工程技术

IEEE Transactions on Intelligent Vehicles Pub Date : 2024-09-06 DOI: 10.1109/TIV.2024.3455566

Pengkun Zhou;Pengfei Gu;Xu Lyu;Maoyou Liao;Ziyang Meng

引用次数: 0