检索结果-内蒙古大学图书馆

Adaptive graph transformer with future interaction modeling for multi-agent trajectory prediction

KNOWLEDGE-BASED SYSTEMS 2025年 316卷

作者： Chen, Xiaobo Wang, Junyu deng, fuwen Li, Zuoyong Shandong Technol & Business Univ Sch Comp Sci & Technol Yantai 264005 Peoples R China Minjiang Univ Sch Comp & Big Data Fuzhou 350121 Peoples R China

Forecasting the trajectories of traffic agents is essential for autonomous systems such as self-driving cars and social robots to guarantee safety in crowded scenarios. Capturing social interactions between agents and generating informative future features bring great challenges to accurate trajectory prediction. To this end, this paper proposes a novel multi-agent trajectory prediction model called AGTFI based on the adaptive graph transformer and future interaction modeling. First, an adaptive graph transformer (AGT) proficient at extracting node and edge features is introduced to capture the complex social interactions between traffic agents. Moreover, a two-stage prediction approach is devised where the first stage is devoted to generating pre-estimated future motion features by bidirectional corrected GRU (BCGRU) and the second stage further incorporates future social interactions into BCGRU to reduce prediction errors. Quantitative and qualitative evaluations of AGTFI on benchmark datasets, including ETH-UCY, SDD, and INTERACTION demonstrate the effectiveness of our model. Ablation studies are conducted to verify the rationale behind the model components.

关键词： Trajectory prediction Graph transformer Attention mechanism Bidirectional decoding

来源：评论

学校读者我要写书评

暂无评论

A dynamic self-improving ramp metering algorithm based on multi-agent deep reinforcement learning

引用

TRANSPORTATION LETTERS-THE INTERNATIONAL JOURNAL OF TRANSPORTATION RESEARCH 2024年第7期16卷 649-657页

作者： deng, fuwen Jin, Jiandong Shen, Yu Du, Yuchuan Shandong Technol & Business Univ Sch Comp Sci & Technol Yantai Peoples R China Peking Univ Sch Elect Engn & Comp Sci Beijing Peoples R China Tongji Univ Coll Transportat Engn Shanghai Peoples R China Shandong Technol & Business Univ Sch Comp Sci & Technol Yantai 264005 Peoples R China

We present a novel ramp metering algorithm that incorporates multi-agent deep reinforcement learning (DRL) techniques, which utilizes monitoring data from loop detectors. Our proposed approach employed a multi-agent DRL framework to generate optimized ramp metering schedules for each ramp meter in real-time, enhancing the operational efficiency of urban freeways with less investment. To simplify the implementation and training of the algorithm, we developed a simulation platform based on SUMO microscopic traffic simulator. We conducted a series of simulation experiments, including local and coordinated ramp metering scenarios with various traffic demands profiles. The simulation results indicate that the proposed DRL-based algorithm outperforms the state-of-the-practice ramp metering methods, considering a comprehensive evaluation index encompassing mainstream speed at the bottleneck and queue length on ramp. Additionally, the method exhibits robustness, scalability, and the potential for further improvement through online learning during implementation.

关键词： ramp metering deep reinforcement learning adaptive control

来源：评论

学校读者我要写书评

暂无评论

Composite Nonconvex Low-Rank Tensor Completion With Joint Structural Regression for Traffic Sensor Networks Data Recovery

引用

IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS 2024年第5期11卷 6882-6896页

作者： Chen, Xiaobo Wang, Kaiyuan Zhao, Feng deng, fuwen Ye, Qiaolin Shandong Technol & Business Univ Sch Comp Sci & Technol Yantai 264005 Peoples R China Nanjing Forestry Univ Coll Informat Sci & Technol Nanjing 210037 Peoples R China Nanjing Forestry Univ State Key Lab Tree Genet & Breeding Nanjing 210037 Peoples R China

Traffic sensor networks allow convenient collection of travel data that are of great significance for intelligent transportation systems (ITSs). However, the universality of missing data impedes the application of ITS and thus accurate missing data recovery is indispensable in practice. Typically, the global low-rankness and local spatiotemporal smoothness exist in underlying traffic tensor data. In light of this, this article proposes an improved low-rank tensor completion (LRTC) model by exploiting abundant structural information from incomplete tensors. Specifically, a logarithm power composite (LPC)-norm is first proposed as a nonconvex substitute of the rank function, leading to a flexible characterization of tensor multidimensional correlation. Then, a joint structural regression (JSR) model is presented to simultaneously leverage the intrinsic temporal continuity and profile similarity of traffic data. By doing so, we construct a novel nonconvex LRTC model by integrating the global low-rankness and fine-grained spatiotemporal structure that are complementary to each other. To solve the proposed model, following the optimization framework of the alternating direction method of multipliers (ADMMs), we develop an efficient iterative algorithm where each step can be solved in a closed form. Extensive experiments on four real-world traffic data are conducted to evaluate the effectiveness of the proposed approach. The results demonstrate that compared with other tensor completion methods, our model significantly improves the recovery performance.

关键词： Tensors Data models Correlation Spatiotemporal phenomena Probabilistic logic Computational modeling Accuracy Nonconvex surrogate spatial-temporal correlation tensor completion traffic sensor networks

来源：评论

学校读者我要写书评

暂无评论

Stochastic Non-Autoregressive Transformer-Based Multi-Modal Pedestrian Trajectory Prediction for Intelligent Vehicles

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2024年第5期25卷 3561-3574页

作者： Chen, Xiaobo Zhang, Huanjia deng, fuwen Liang, Jun Yang, Jian Shandong Technol & Business Univ Sch Comp Sci & Technol Yantai 264005 Shandong Peoples R China Jiangsu Univ Automot Engn Res Inst Zhenjiang 212013 Jiangsu Peoples R China Nankai Univ Coll Comp Sci Tianjin Key Lab Visual Comp & Intelligent Percept Tianjin 300350 Peoples R China

Pedestrian trajectory prediction, which aims at predicting the future positions of all pedestrians in a crowd scene given their past trajectories, is the cornerstone of autonomous driving and intelligent transportation systems. Accurate prediction and fast inference are both indispensable for real-world applications. In this paper, we propose a stochastic non-autoregressive Transformer-based multi-modal trajectory prediction model to address the two challenges. Specifically, a novel graph attention module dedicated to joint learning of social and temporal interaction is proposed to explore the complex interaction among pedestrians while integrating sparse attention mechanism, pedestrian identity, and temporal order contained in the trajectory data. By doing so, the interaction across temporal and social dimensions can be simultaneously processed to extract abundant context features for prediction. Besides, to accelerate inference speed, we put forward a stochastic non-autoregressive Transformer model with multi-modal prediction capability where each future trajectory can be inferred in a parallel fashion, therefore, resulting in diverse trajectory predictions and less computational cost. Extensive experiments and ablation studies are performed to evaluate our approach. The empirical results demonstrate that the proposed model not only produces high prediction accuracy but also infers with fast speed. The code of the proposed method will be publicly available at https://***/xbchen82/SNARTF.

关键词： Pedestrian trajectory prediction graph attention non-autoregressive transformer multi-modal prediction

来源：评论

学校读者我要写书评

暂无评论

Transferable driver facial expression recognition based on joint discriminative correlation alignment network with enhanced feature attention

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2023年第12期17卷 2444-2457页

作者： Chen, Xiaobo Du, Jian deng, fuwen Zhao, Feng Shandong Technol & Business Univ Sch Comp Sci & Technol Yantai Shandong Peoples R China

Facial expression is closely related to the emotions of drivers, thus facilitating safe driving detection in advanced driving assistance system (ADAS). Recently, deep learning techniques have become prevalent for facial expression recognition. However, for driving scenarios, the facial expression recognition is mainly challenged by the problems of small sample size as well as effective feature representation. To address the above issues, this paper puts forward a transfer learning-based method for driver facial expression recognition by fully exploiting the other sources of facial expression data that may obey different distributions. Specifically, an enhanced feature attention module is firstly devised such that affluent features with multi-scales can be extracted and refined based on the spatial and channel attention mechanisms. Then, a joint correlation alignment loss is presented ensuring that the samples in the source and target domains are transformed into the shared common subspace to reduce difference of both the marginal and conditional distributions. Multiple transfer learning tasks on real-world data are carried out to evaluate the proposed method. The experimental results show that our model achieves better recognition accuracy for driver facial expression compared with several traditional and deep learning-based transfer learning algorithms.

关键词： correlation alignment driver expression recognition spatial-channel attention transfer learning

来源：评论

学校读者我要写书评

暂无评论

Practical Applications of ChatGPT and BERT: Long Text Summarization and Model Feature Ablation Study for Stock Comment Topic Identification 24

Practical Applications of ChatGPT and BERT: Long Text Summar...

引用

2024 Guangdong-Hong Kong-Macao Greater Bay Area International Conference on Digital Economy and Artificial Intelligence, DEAI 2024

作者： Xu, Mingyan Chen, Hongyu deng, fuwen Liu, Yuan Zhao, Feng Zhang, Yong School of Statistics Shandong Technology and Business University Yantai264000 China School of Hospital Shandong Technology and Business University Yantai264000 China School of Computer Science and Technology Shandong Technology and Business University Yantai264000 China Yantai Key Laboratory of Big Data Modeling and Intelligent Computing School of Computer Science and Technology Shandong Technology and Business University Yantai China School of Information and Electrotic Shandong Technology and Business University Yantai264000 China

ISBN: (纸本)9798400717147

Topic identification of expert stock comments is an important task in the field of finance, which helps investors understand the market dynamics and expert opinions so that they can make more informed investment decisions. The Bidirectional Encoder Representations from Transformers (BERT) model is one of the widely used methods for text topic identification. To improve the identification accuracy and overcome the limitations of the BERT model, two strategies are proposed in this study. First, given the BERT model's input length limitation when dealing with long texts, we used the summaries created by Chat Generative Pre-training Transformer (ChatGPT) as inputs to the model. The generative summarization capability of ChatGPT not only reduces the length of the text effectively but also preserves the key information of the original text, which enables the model to better understand and deal with long texts. Second, we perform multi-layer feature ablation on the BERT model to find the most representative features for text topic identification. The multi-layer features of the BERT model can be more effectively utilized to improve the accuracy of text topic identification. Experiments show that our proposed strategies exhibit excellent performance and provide effective paths to improve the accuracy of stock comment topic identification. © 2024 ACM.

关键词： Ablation

来源：评论

学校读者我要写书评

暂无评论

Enhancing automated health monitoring of road infrastructure through a hierarchical robust deep learning approach

引用

STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL 2024年

作者： deng, fuwen Jin, Jiandong Chen, Xiaobo An, Zhiyong Du, Yuchuan Shandong Technol & Business Univ Sch Comp Sci & Technol 191 Mid Binhai Rd Yantai 264005 Shandong Peoples R China Peking Univ Sch Elect Engn & Comp Sci Beijing Peoples R China Tongji Univ Coll Transportat Engn Shanghai Peoples R China

The life-cycle monitoring of road health conditions is crucial for precise and automated maintenance of road infrastructure. Despite this importance, existing machine vision-based methods for identifying pavement distresses are constrained in their susceptibleness to interference from factors such as shadows. To bridge this gap, the article introduces a novel hierarchical pavement structure monitoring model, which significantly enhances the task's performance. This proposed model integrates a de-shadowing module based on generative adversarial networks and an anchor-free instance segmentation model to provide pixel-level detection results, encompassing distress categories and morphological masks. A large-scale pavement image dataset was constructed, and a generative model was developed to expand pavement shadow samples. Experimental results show that the model achieved an average precision metric of 78.2%, surpassing existing baseline models. The effectiveness of the de-shadowing module was also validated through the significant improvements observed in various evaluation indicators. The proposed method demonstrates superior accuracy and robustness, holding promise for real-world road infrastructure monitoring systems.

关键词： Pavement health monitoring robustness enhancement hierarchical deep learning instance segmentation model

来源：评论

学校读者我要写书评

暂无评论

VOCs Concentration, SOA Formation Contribution and Festival Effects during Heavy Haze Event: A Case Study in Zhengzhou, Central China

引用

ATMOSPHERE 2024年第8期15卷 1009页

作者： Yu, Shijie Xue, Chaofang deng, fuwen Xu, Qixiang Zhao, Bingnan Henan Univ Sci & Technol Dept Environm Engn Luoyang 471023 Peoples R China Zhengzhou Univ Inst Environm Sci Zhengzhou 450001 Peoples R China Zhengzhou Univ Sch Ecol & Environm Zhengzhou 450001 Peoples R China

In this study, online ambient volatile organic compounds (VOCs) were collected at an urban site of Zhengzhou in Central China during February 2018. The VOCs characteristics, source contributions and the Chinese New Year (CNY) effects have been investigated. During the sampling period, three haze periods have been identified, with the corresponding VOCs concentrations of (92 +/- 45) ppbv, (62 +/- 18) ppbv and (83 +/- 34) ppbv;in contrast, the concentration during non-haze days was found to be (57 +/- 27) ppbv. In addition, the festival effects of the CNY were investigated, and the concentration of particulate matter precursor decreased significantly. Meanwhile, firework-displaying events were identified, as the emission intensity had been greatly changed. Both potential source contribution function (PSCF) and the concentration weighted trajectory (CWT) models results indicated that short-distance transportation was the main influencing factor of the local VOCs pollution, especially by transport from the northeast. Source contribution results by the positive matrix factorization (PMF) model showed that vehicle exhaust (24%), liquid petroleum gas and natural gas (LPG/NG, 23%), coal combustion (21%), industrial processes (16%) and solvent usages (16%) were the major sources of ambient VOCs. Although industry and solvents have low contribution to the total VOCs, their secondary organic aerosol (SOA) contribution were found to be relatively high, especially in haze-1 and haze-3 periods. The haze-2 period had the lowest secondary organic aerosol potential (SOAp) during the sampling period;this is mainly caused by the reduction of industrial and solvent emissions due to CNY.

关键词： volatile organic compounds (VOCs) heavy haze source apportionment secondary organic aerosols potential (SOAp) Chinese New Year (CNY) festival effects

来源：评论

学校读者我要写书评

暂无评论

On the scheduling of lifecycle network-level pavement maintenance: modelling and applications

引用

INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING 2023年第2期24卷

作者： deng, fuwen Jin, Jiandong Chen, Xiaobo Zhao, Feng Shandong Technol & Business Univ Sch Comp Sci & Technol Yantai Peoples R China Peking Univ Sch Elect Engn & Comp Sci Beijing Peoples R China

Pavement maintenance is a prominent issue in road traffic systems and plays a fundamental role in ensuring the safety, mobility and comfort of road networks worldwide. This paper introduces an innovative approach for scheduling network-level pavement maintenance based on mathematical planning, guided by the principle of lifecycle cost analysis (LCCA). The primary objective is to optimise pavement performance while minimising maintenance costs, ultimately achieving cost-effectiveness. In particular, the method aims to solve the '2W1H' problem (i.e. 'Which', 'When', and 'How'), which implies identifying the specific road sections, determining an appropriate maintenance timeline, and selecting the most effective maintenance treatment for long-term pavement maintenance planning. The proposed method comprises a bi-level optimisation model. The upper layer is a multi-objective optimisation model which concurrently maximising the lifecycle pavement performance while minimising the lifecycle costs;the lower layer evaluates the impact of pavement maintenance operations on travellers' route choice behaviour by employing the user equilibrium problem to determine the change in user cost attributable to a specific maintenance scheduling scheme. A meta-heuristic algorithm is utilised to solve the model and acquire a Pareto front for multi-objective optimisation. Furthermore, we discuss the practical engineering applications under budgetary constraints and pavement performance requirements, followed by the consideration of parameter uncertainties in pavement performance prediction models. The numerical experiments demonstrate that our approach can deliver pavement maintenance scheduling outcomes across various scales of road networks, and highlight the promising potential of our approach for future pavement management endeavours.

关键词： Lifecycle pavement maintenance multi-objective optimisation meta-heuristic algorithm

来源：评论

学校读者我要写书评

暂无评论

Pyrolytic Graphite for an In-Plane Force Study of Diamagnetic Levitation: A Potential Microdetector of Cracks in Magnetic Material

引用

MICROMACHINES 2023年第6期14卷 1242-1242页

作者： Liu, Runze Yang, Wenjiang Xiang, Hongjun Zhao, Peng deng, fuwen Yan, Juzhuang Beihang Univ Sch Astronaut Beijing 100191 Peoples R China

The diamagnetic levitation technique can be applied in non-destructive testing for identifying cracks and defects in magnetic materials. Pyrolytic graphite is a material that can be leveraged in micromachines due to its no-power diamagnetic levitation on a permanent magnet (PM) array. However, the damping force applied to pyrolytic graphite prevents it from maintaining continuous motion along the PM array. This study investigated the diamagnetic levitation process of pyrolytic graphite on a permanent magnet array from various aspects and drew several important conclusions. Firstly, the intersection points on the permanent magnet array had the lowest potential energy and validated the stable levitation of pyrolytic graphite on these points. Secondly, the force exerted on the pyrolytic graphite during in-plane motion was at the micronewton level. The magnitude of the in-plane force and the stable time of the pyrolytic graphite were related to the size ratio between it and the PM. During the fixed-axis rotation process, the friction coefficient and friction force decreased as the rotational speed decreased. Smaller-sized pyrolytic graphite can be used for magnetic detection, precise positioning and other microdevices. The diamagnetic levitation of pyrolytic graphite can also be used for detecting cracks and defects in magnetic materials. We hope this technique will be used in crack detection, magnetic detection and other micromachines.

关键词： pyrolytic graphite diamagnetic levitation cracks detector in-plane motion fixed-axis rotation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：