检索结果-内蒙古大学图书馆

Efficient and High-Quality Path Planning Guided by Map Information for Mobile Robots

IEEE-ASME TRANSACTIONS ON MECHATRONICS 2025年

作者： Fan, Wenpei Wang, Yaonan Liu, Licheng Chen, Yanjie Tan, Haoran Miao, Zhiqiang Hunan Univ Coll Elect & Informat Engn Changsha 410082 Peoples R China Hunan Univ Natl Engn Res Ctr Robot Visual Percept & Control T Changsha 410082 Peoples R China Fuzhou Univ Sch Mech Engn & Automat Fuzhou 350108 Peoples R China Aberystwyth Univ Dept Comp Sci Aberystwyth SY23 3DB Wales

Path planning of mobile robots plays an important role in improving work efficiency in the manufacturing industry. On one hand, the sampling-based path planning methods commonly used in mobile robots require extensive and time-consuming iterative searches to find a relatively optimal path due to their stochastic nature. On the other hand, the map information obtained during planning is solely utilized for collision detection without being effectively leveraged. This article proposes a map-informed rapidly random-exploring tree (MIRRT) algorithm that combines uniform sampling, rapidly crossing obstacles (RCO) and progressive optimization. It can utilize map information to quickly guide the search tree. MIRRT involves a preprocessing step that requires segmenting concave obstacles into convex obstacles to prevent path loss. During the growth of the search tree, the RCO method guided by map information can rapidly discover high-quality ground path nodes and efficiently navigate obstacles when they block the path. Finally, simulations and physical experiments were performed to prove that MIRRT significantly outperforms existing state-of-the-art sampling-based planners designed specifically for mobile robots.

关键词： Path planning Planning Mobile robots classification algorithms Robot sensing systems Heuristic algorithms Hands Deep learning Convergence Fans Dynamic planning map information mobile robots path planning rapidly exploring random tree

来源：评论

学校读者我要写书评

暂无评论

UCBEE: A Multi Armed Bandit Approach for Early-Exit in Neural Networks

引用

IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT 2025年第1期22卷 107-120页

作者： Pacheco, Roberto G. Bajpai, Divya J. Shifrin, Mark Couto, Rodrigo S. Menasche, Daniel Sadoc Hanawal, Manjesh K. Campista, Miguel Elias M. Univ Fed Fluminense Comp Sci Dept BR-28895532 Rio das Ostras Brazil Indian Inst Technol Dept Ind Engn & Operat Res Mumbai 400076 India Ben Gurion Univ Negev Sch Elect & Comp Engn IL-8410501 Beer Sheva Israel Univ Fed Rio de Janeiro Elect & Comp Engn Dept BR-21941901 Rio De Janeiro Brazil Univ Fed Rio de Janeiro Inst Comp BR-21941901 Rio De Janeiro Brazil

Deep Neural Networks (DNNs) have demonstrated exceptional performance in diverse tasks. However, deploying DNNs on resource-constrained devices presents challenges due to energy consumption and delay overheads. To mitigate these issues, early-exit DNNs (EE-DNNs) incorporate exit branches within intermediate layers to enable early inferences. These branches estimate prediction confidence and employ a fixed threshold to determine early termination. Nonetheless, fixed thresholds yield suboptimal performance in dynamic contexts, where context refers to distortions caused by environmental conditions, in image classification, or variations in input distribution due to concept drift, in NLP. In this article, we introduce Upper Confidence Bound in EE-DNNs (UCBEE), an online algorithm that dynamically adjusts early exit thresholds based on context. UCBEE leverages confidence levels at intermediate layers and learns without the need for true labels. Through extensive experiments in image classification and NLP, we demonstrate that UCBEE achieves logarithmic regret, converging after just a few thousand observations across multiple contexts. We evaluate UCBEE for image classification and text mining. In the latter, we show that UCBEE can reduce cumulative regret and lower latency by approximately 10%-20% without compromising accuracy when compared to fixed threshold alternatives. Our findings highlight UCBEE as an effective method for enhancing EE-DNN efficiency.

关键词： Image classification Image edge detection Distortion Accuracy Performance evaluation classification algorithms Delays Proposals Neural networks Natural language processing Multi armed bandits early-exit natural language processing image classification

来源：评论

学校读者我要写书评

暂无评论

Travel mode choice prediction: developing new techniques to prioritize variables and interpret black-box machine learning techniques

引用

TRANSPORTATION PLANNING AND TECHNOLOGY 2025年第3期48卷 582-605页

作者： Naseri, Hamed Waygood, E. O. D. Patterson, Zachary Alousi-Jones, Meredith Wang, Bobin Polytech Montreal Dept Civil Geol & Min Engn Montreal PQ H3T 1J4 Canada Concordia Univ Concordia Inst Informat Syst Engn Montreal PQ Canada Univ Laval Dept Mech Engn Quebec City PQ Canada

Travel Mode Choice (TMC) prediction is vital for forecasting travel demand and transportation planning. To be helpful for those purposes, one needs to know with high accuracy what influences choices and how. For accuracy, Machine Learning (ML) classification techniques often produce results with higher accuracy than traditional methods. However, many ML techniques are black-box tools, making them less useful for planning. To this end, two new approaches were proposed to interpret the results of ML techniques and investigate the influence of different variables on TMC. The results suggested that ensemble learning techniques outperform other prediction methods. Adding accessibility, geographic, and land-use variables to the conventional TMC prediction models could improve their performance. The most important parameters for TMC were found to be: trip distance, availability of a transit pass and availability of a driver's license. Their respective influences on the different modes are demonstrated using the novel method mentioned above.

关键词： Travel mode choice classification algorithms feature analysis gray relational analysis

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Techniques for Estimating Energy Expenditure in Wheelchair Users

引用

IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING 2025年 33卷 739-749页

作者： Doshmanziari, Roya Aandahl, Hakon Strand Reierstad, Havard Pettersen Danielsson, Marius Lyng Baumgart, Julia Kathrin Varagnolo, Damiano Norwegian Univ Sci & Technol Dept Engn Cybernet N-7034 Trondheim Norway Nord Univ Fac Teacher Educ & Arts Dept Phys Educ & Sport Sci N-7600 Levanger Norway Norwegian Univ Sci & Technol Dept Neuromed & Movement Sci Ctr Elite Sports Res N-7034 Trondheim Norway Univ Padua Dept Informat Engn I-35122 Padua Italy

Providing feedback on energy expenditure (EE) may be an important tool to support obesity prevention among manual wheelchair users (MWU). This paper presents a data-driven approach for estimating EE based on data collected from 40 participants (20 MWU and 20 controls without disability) across different activities (lying, sitting and wheelchair propulsion at different intensities). We extracted features from heart rate, inertial measurement units (IMU), and individual personal characteristics to develop activity classification and EE estimation algorithms and investigate the influence of personal characteristics on EE estimates. Support Vector Machines were selected as classifiers, while Support Vector Regressors, Gaussian Processes, Random Forest, XGBoost, and Neural Networks were selected as regression models. High classification accuracy was achieved with minor confusion between activities and EE estimation results showed high generalisation capabilities of the trained models on unseen participants. We explored the impact of changing the position of the IMU on the accuracy of EE estimations. We recommend the wrist as the primary location for sensor placement. It provides a good trade-off between accuracy, high wear compliance rates and the possibility of integrating our algorithms in already existing wearable devices. Our findings showed that including data collected from people without disabilities to develop EE estimation algorithms for MWU did not enhance the estimation accuracy. In conclusion, data-driven algorithms based on wearable sensors and personal characteristics are effective for activity classification and EE estimation in MWU, but need to be personalized and further developed for daily life settings to be ecologically valid.

关键词： Feature extraction Wheelchairs Estimation Accuracy Gyroscopes Propulsion Heart rate classification algorithms Wrist Wearable sensors Manual wheelchair users energy expenditure inertial measurement units wearable sensors

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Driven Optimization for Solution Space Reduction in the Quadratic Multiple Knapsack Problem

引用

IEEE ACCESS 2025年 13卷 10638-10652页

作者： Yanez-Oyarce, Diego Contreras-Bolton, Carlos Troncoso-Espinosa, Fredy Rey, Carlos Univ Bio Bio Dept Ingn Ind Concepcion 3780000 Chile Univ Concepcion Dept Ingn Ind Concepcion 4070409 Chile

The quadratic multiple knapsack problem (QMKP) is a well-studied problem in operations research. This problem involves selecting a subset of items that maximizes the linear and quadratic profit without exceeding a set of capacities for each knapsack. While its solution using metaheuristics has been explored, exact approaches have recently been investigated. One way to improve the performance of these exact approaches is by reducing the solution space in different instances, considering the properties of the items in the context of QMKP. In this paper, machine learning (ML) models are employed to support an exact optimization solver by predicting the inclusion of items with a certain level of confidence and classifying them. This approach reduces the solution space for exact solvers, allowing them to tackle more manageable problems. The methodological process is detailed, in which ML models are generated and the best one is selected to be used as a preprocessing approach. Finally, we conduct comparison experiments, demonstrating that using a ML model is highly beneficial for reducing computing times and achieving rapid convergence.

关键词： classification algorithms Prediction algorithms Metaheuristics Genetic algorithms Synthetic data Heuristic algorithms Correlation Support vector machines Standards Mathematical models Machine learning combinatorial optimization knapsack problem quadratic multiple knapsack problem

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Machine Learning Model for Efficient XML Parsing

引用

IEEE ACCESS 2025年 13卷 382-393页

作者： Ali, Muhammad Khan, Minhaj Ahmad Ur Rasool, Raihan Bahauddin Zakariya Univ Dept Comp Sci Multan 60800 Pakistan IBM Technol Melbourne Vic 3006 Australia

The Extensible Markup Language (XML) files are extensively used for representing structured data on the web for file configuration, exchanging data between distinct applications, web development, and many other applications. Consequently, effective parsing techniques are necessary for XML files to enhance the performance of applications. The existing parsing techniques have their strengths and weaknesses affecting the performance of applications. Researchers point out that the selection of an efficient and appropriate parser is the most challenging issue regarding a particular condition. This paper proposes a framework XML Parsing Optimization using Hybrid Machine Learning (XPOHML) that makes use of Artificial Neural Network (ANN) and Support Vector Machine (SVM) machine learning techniques for efficient XML parsing. The newly developed framework performs analysis and prediction of different XML parsers using profiling, classification, performance evaluation, and finally generates code for efficient parsing. The XML profiling phase of the XPOHML framework generates a dataset by evaluating the performance of PXTG, SAX, StAX, DOM, and JDOM parsing models on separate cores by applying numerous file sizes. The classification phase produces the classification model by applying ANN and SVM techniques to identify the appropriate parsing model. The performance evaluation phase of XPOHML assesses the performance of both parsing models through classification metrics (accuracy). Additionally, based on evaluation outcomes, the code generation phase produces an efficient parsing model of XML files. The newly designed and developed XPOHML framework has shown a meaningful improvement in the performance of parsing XML files.

关键词： XML Support vector machines classification algorithms Prediction algorithms Machine learning Codes Machine learning algorithms Predictive models Performance evaluation Training Artificial intelligence artificial neural network eXtensible markup language framework machine learning model parsing support vector machine XML parsing optimization

来源：评论

学校读者我要写书评

暂无评论

FWA-SVM Network Intrusion Identification Technology for Network Security

引用

IEEE ACCESS 2025年 13卷 18579-18593页

作者： Zhang, Yaohui Hunan Post & Telecommun Coll Informat & Commun Dept Changsha 410000 Hunan Peoples R China

In the digital age, the increasing demand for network security has driven research on efficient network intrusion detection systems. The effectiveness of traditional network intrusion is limited in the face of complex network attacks and constantly increasing data volume. Support vector machine has attracted much attention for its excellent classification ability, but it faces challenges in feature selection and parameter optimization when dealing with large-scale high-dimensional data. Therefore, the study introduces the fireworks algorithm to improve it and optimize parameter selection and feature subset selection. The study also proposes a discretized binary fireworks algorithm to further improve the efficiency and adaptability of support vector machines in feature selection. The experiment outcomes denote that on the feature dense Sonar dataset, the average number of features selected by the proposed method is 24.39, a decrease of 25.51% compared to the comparison algorithm, and a classification accuracy improvement of 2.99%. The average detection rate of the raised method is 96.43%, the false alarm rate is 0.91, and the average correlation coefficient is as high as 0.987, which is better than the other four comparative algorithms. The training and testing time are 26.12 seconds and 11.23 seconds, respectively. Consequently, the primary contribution of the research lies in solving key problems in network intrusion detection through the BFWA-SVM model. Introducing the discrete fireworks algorithm to optimize support vector machines has improved the ability to process large-scale high-dimensional data. This model significantly reduces false positive and false negative rates, enhances real-time security awareness, and provides new guidance for network security management.

关键词： Support vector machines Optimization Feature extraction classification algorithms Accuracy Network security Fireworks algorithm Adaptation models Convolutional neural networks High dimensional data FWA SVM network security intrusion detection technology binary encoding feature selection

来源：评论

学校读者我要写书评

暂无评论

A Survey on Adversarial Attacks for Malware Analysis

引用

IEEE ACCESS 2025年 13卷 428-459页

作者： Aryal, Kshitiz Gupta, Maanak Abdelsalam, Mahmoud Kunwar, Pradip Thuraisingham, Bhavani Tennessee Technol Univ Dept Comp Sci Cookeville TN 38505 USA North Carolina A&T State Univ Dept Comp Sci Greensboro NC 27411 USA Univ Texas Dallas Dept Comp Sci Dallas TX 75080 USA

Machine learning-based malware analysis approaches are widely researched and deployed in critical infrastructures for detecting and classifying evasive and growing malware threats. However, minor perturbations or ineffectual byte insertions can easily 'fool' these trained ML classifiers, making them ineffective against these crafted and smart malicious software. This survey aims to provide an encyclopedic overview of adversarial evasion attacks specifically targeting malware detection and classification systems, standing apart from previous surveys by focusing exclusively and comprehensively on this unique application domain. While significant strides have been made in adversarial research in other fields, the specific challenges of adversarial malware remain under-explored due to the intricate nature and constraints of the malware domain. Our survey addresses this gap by analyzing literature on adversarial evasion attacks published between 2013 and 2024, making it one of the first to systematically focus on malware-specific adversarial attacks in a detailed, self-contained manner. The paper will begin by introducing various machine-learning techniques used to generate adversarial malware samples, including the structural nuances of target files, which influence adversarial vulnerabilities. The work presents an in-depth threat model specific to adversarial malware evasion attacks, describing the unique attack surfaces of malware detectors and outlining adversarial goals tailored to the malware domain. We systematically analyze adversarial generation algorithms from broader domains adapted to malware evasion attacks, proposing a taxonomy of adversarial evasion attacks within malware detection based on target domains(Windows, Android and PDF). The survey highlights real-world adversarial evasion attacks on machine learning-based anti-malware engines under each taxonomical heading, demonstrating the evolution and refinement of these attack strategies over time. Our sur

关键词： Malware Surveys Security Perturbation methods Glass box classification algorithms Engines Detectors Complexity theory Training data Adversarial evasion attack adversary modeling security for AI windows PE malware Android malware PDF malware

来源：评论

学校读者我要写书评

暂无评论

RP-CFANet: An Adaptive Photovoltaic Hot-Spot Fault Detection Network Based on Region Perception and Cross-Channel Feature Aggregation

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Hao, Shuai Qi, Tianrui Ma, Xu Li, Jiahao Li, Tianqi Xian Univ Sci & Technol Coll Elect & Control Engn Xian 710054 Peoples R China

In the process of utilizing thermal infrared sensors to identify photovoltaic (PV) hot-spot faults, due to the diverse shapes of hot-spot faults and environmental interferences, fault characteristics cannot be effectively expressed, which poses a challenge for traditional detection networks in achieving accurate detection. Consequently, an adaptive PV hot-spot fault detection network based on region perception and cross-channel feature aggregation is proposed. First, to reduce the interference of pseudo-highlight features in complex backgrounds, a lite-UNet segmentation network is designed to remove background redundant information and enable the detection network to concentrate on the region of the PV panel. Second, to fully capture the geometric deformations and weak edge characteristics of hot-spot faults, a C2f_DCN module is designed, which enhances the feature extraction capabilities by adaptively adjusting the receptive field. Subsequently, to address the problem of feature coupling often encountered in extracting dense hot-spot fault features, a dense object visual enhancement (DOVE) module is proposed. Through the dynamic aggregation of cross-space features, the feature information of different channels is integrated to improve the detection accuracy of hot-spot faults. Additionally, to further enhance the effective fusion of multiscale features, an adaptive scale converter module (ASCM) detection head is designed. Finally, seven traditional detection methods are chosen for comparison to confirm the benefits of the proposed method. According to the experimental results, the suggested method can detect hot-spot faults with an accuracy of up to 86.5% in complicated inspection environments, which is an increase of 5.6% compared with the basis of the original network.

关键词： Feature extraction Accuracy Fault detection Data mining Fault diagnosis Shape Heuristic algorithms classification algorithms Autonomous aerial vehicles Attention mechanisms Attention mechanism cross-channel feature aggregation hot-spot fault detection infrared sensors U-Net

来源：评论

学校读者我要写书评

暂无评论

A Multitask Deep Learning Approach for sEMG-Based Human Motion Intention and Muscle Fatigue Levels Recognition

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Tu, Pengjia Li, Junhuai Wang, Huaijun Xian Univ Sci & Technol Coll Comp Sci & Technol Xian 710054 Peoples R China Xian Univ Technol Sch Comp Sci & Engn Xian 710048 Peoples R China

The natural interaction between the lower limb prosthetic and amputee directly affects the rehabilitation effect. Human motion intention and muscle fatigue are the most relevant factors affecting the rehabilitation training of amputees. Previous methods focus on structuring independent models for each task, neglecting the problem of motion intention recognition (MIR) under muscle fatigue constraints. To remedy this deficiency, this article proposes a multitask deep learning approach to tackle MIR and muscle fatigue classification (MFC) tasks simultaneously. First, the surface electromyography (sEMG) samples dataset of three limb motions with different muscle fatigue levels are collected and preprocessed. Second, we create a MIR recognition model based on neural networks and then utilize its knowledge-sharing backbone feature network to learn a new MFC task with a limited dataset. Next, a multitask loss function with uncertainty is leveraged to weigh different losses between two tasks and improve model performance. Finally, we compare the accuracy of the proposed model against other models train individually on each task, demonstrating the proposed model effectiveness in terms of data efficiency. Simultaneously, the proposed approach achieves 96.8% accuracy in MIR task under fatigue constraints, while achieves 93.2% accuracy in MFC task which a limited dataset, respectively.

关键词： Muscles Fatigue Limbs Accuracy Training Exoskeletons Feature extraction classification algorithms Data models Legged locomotion Knowledge-sharing backbone feature network motion intention multitask learning muscle fatigue constraint surface electromyography (sEMG)

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：