检索结果-内蒙古大学图书馆

International Conference on control, Artificial Intelligence, Robotics & Optimization (ICCAIRO)

作者： Aristeidis Geladaris Lampis Papakostas Athanasios Mastrogeorgiou Michael Sfakiotakis Panagiotis Polygerinos TWI-Hellas Athens Greece Mechanical Engineering Dept. Control Systems and Robotics Lab (CSRL) School of Engineering Hellenic Mediterranean University Heraklion Crete Greece Control Systems Lab (CSL) School of Mechanical Engineering National Technical University of Athens Greece Electrical and Computer Engineering Dept. Control Systems and Robotics Lab (CSRL) School of Engineering Hellenic Mediterranean University Heraklion Crete Greece

In this paper we propose a method for autonomous navigation in GPS-denied environments that can be deployed in all types of robotic vehicles requiring only a single RGB-D camera. Our algorithm is an integration of localization, mapping and trajectory planning software modules that can handle dynamic environments. dept. measurements and visual odometry are used to (a) create a robocentric Euclidean Signed Distance Fields (ESDF) map in real-time, (b) estimate the position of obstacles surrounding the robot, and (c) calculate the distance and gradient from them. Subsequently, we use an optimization algorithm to plan a collision-free trajectory. We show that a local map can be created faster than with the fixed-size array method used in current optimization methods. Our approach also allows for effective detection of dynamic obstacles by constantly updating the map. As a result, the robot deviates from its initial path only when necessary. We validate our results in simulation by deploying our algorithm in a skid steering rover and a hexacopter in an environment with static and dynamic obstacles.

关键词：

来源：评论

学校读者我要写书评

暂无评论

TCCFD: An Efficient Tree-based Framework for Credit Card Fraud Detection 16

TCCFD: An Efficient Tree-based Framework for Credit Card Fra...

引用

16th International Conference on computer engineering and systems, ICCES 2021

作者： Own, Reem M. Salem, Sameh A. Mohamed, Amr E. Helwan University Faculty of Engineering Computer Systems Engineering Dept. Cairo Egypt

ISBN: (纸本)9781665408677

Due to digitalization, credit card fraud has become one of the most prevalent global risks. The global economy loses billions of dollars annually due to credit card fraud. Financial institutions adopt a strategic approach to credit card fraud mitigation, relying on business rules to combat the occurrence of fraud and sophisticated analysis to detect fraud. However, Fraudsters are altering their patterns regularly with the help of new technologies. Thus, organizations continually seek to detect new fraud patterns. The detection of credit card fraud has improved tremendously through using artificial intelligence, particularly machine learning techniques. We introduce through this paper an efficient framework that adopts various tree-based machine learning models to detect fraudulent transactions. We apply the proposed framework to a highly asymmetric real-world dataset. The proposed framework integrating the extra trees classifier with the borderline synthetic minority oversampling technique surpasses others and previously developed models with 99.96 % accuracy and 96% area under the ROC curve. The proposed framework asserts its efficacy and reliability in classification while also addressing the class imbalance issue. © 2021 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

BioPoint: Enhancing Human-computer Interaction through Single-Site, Multi-Sensor Gesture Recognition

BioPoint: Enhancing Human-Computer Interaction through Singl...

引用

Annual International Conference of the IEEE engineering in Medicine and Biology Society (EMBC)

作者： Xavier Isabel Evan Campbell Loïc Olivier Gabriel Gagné Erik Scheme Ulysse Côté-Allard Gabriel Gagnon-Turcotte Dept. of Electrical and Computer Engineering Laval University Québec Canada Department of Electrical and Computer Engineering University of New Brunswick Fredericton Canada Dept. of Technology Systems University of Oslo Oslo Norway

ISBN: (数字)9798350371499

ISBN: (纸本)9798350371505

Advancements in human-computer interaction (HCI) and machine learning are seen as key avenues to help individuals living with upper limb disabilities in accomplishing their activities of daily living. Multi-channel myoelectric systems are a promising approach for HCI due to their intuitive and accurate capture of user intent through muscle activity. However, such systems are still bulky compared to widely accepted smartwatches-like devices and as such pose a challenge for seamless integration into daily life. Consequently, this work presents the first single-site EMG-IMU-PPG system for multi-gesture recognition from the wrist. Overall, when evaluated on eight able-bodied participant, the system achieves an average accuracy of 65.28%±10.45% on 17 finger and wrist gestures and 90.98%±5.65% on eight hand gestures. This study also introduces "Myo Hero", an online environment inspired by "Guitar Hero." Myo Hero provides a dynamic method for testing multi-gesture recognition systems, improving the evaluation process’s appeal. Online evaluation of the multimodal system substantially outperformed random play, signifying EMG-IMU-PPG’s potential for human-computer interaction.

关键词： Human computer interaction Wrist Accuracy Pipelines Gesture recognition Rhythm Electromyography Recording Synchronization Testing

来源：评论

学校读者我要写书评

暂无评论

Design of RISE control for Respiratory System 8

Design of RISE Control for Respiratory System

引用

8th IEEE International Conference on engineering Technologies and Applied Sciences, ICETAS 2023

作者： Husain, Suha S. Rasheed, Luay T. Mahmod, Rawnaq A. Hamza, Ekhlas K. Noaman, Noaman M. Humaidi, Amjad Jaleel University of Technology Department of Construction and Projects Baghdad Iraq University of Technology Control and Systems Eng. Department Baghdad Iraq Training & Energy Researches Office Ministry of Electricity Baghdad Iraq University of Technology Department of Mechatronics Engineering Bahrain University of Technology Control and Systems Eng. Dept. Baghdad Iraq

ISBN: (纸本)9798350327090

In this study, a new control strategy is presented and proposed, which is the Robust Integral of the Sign of Error (RISE). The RISE control strategy is designed to track the airway pressure in the mechanical ventilate process. The respiratory model is designed by combining a blower hose model and a single compartmental lung system with nonlinear lung compliance. For comparison purposes, the Sliding Mode control (SMC) strategy is designed and compared to the proposed control mechanism. RISE controller generates a smooth function;therefore, RISE is able to avoid the chattering phenomenon caused by the discontinuous part of the SMC. In addition, RISE control provides better performance than SMC and fewer tracking error. © 2023 IEEE.

关键词： Sliding mode control

来源：评论

学校读者我要写书评

暂无评论

EEG Signal Power Prediction Using DEAP Dataset 7

EEG Signal Power Prediction Using DEAP Dataset

引用

7th International Conference on Intelligent Informatics and Biomedical Sciences, ICIIBMS 2022

作者： Munoz-Gonzalez, Angel Horie, Ryota Shibaura Institute of Technology Functional Control Systems Course Tokyo Japan Shibaura Institute of Technology Dept. of Information and Communications Engineering Tokyo Japan

ISBN: (数字)9781665482295

ISBN: (纸本)9781665482295

When we listen to music, our emotions can change due to changes in our brain response. Because of this, a large number of research projects for classifying the emotional response using brain signals when listening to music through machine learning techniques emerged. On the contrary, to our knowledge, there is no previous research attempting to estimate the dynamic changes in the electroencephalogram (EEG) response under music stimuli through machine learning techniques. Therefore, in this manuscript, we proposed an approach to predict and anticipate changes in the EEG signal under music stimuli. Using the DEAP dataset, we split the EEG response to music stimuli into one-second length frames. After that, we compared the changes in the power of the brain signal of consecutive frames through two one-tailed Wilcoxon rank-sum tests. This test allowed us to label the changes in the second frame as "lower", "similar"or "higher"signal compared to the first frame. Then, we attempted to predict these changes using a Support-Vector Machine (SVM) classifier with stratified 5-fold validation with different input combinations (only music, only brain signal, or a combination of both). Due to the use of multi-label classification with imbalanced data, we measured the results through F1-Scores. Over chance level predictions of the changes of signal power were obtained when using the previous second brain signal for the different channels and bands, especially in the frontal F3 and F4 channels. © 2022 IEEE.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Fast and Accurate LSTM Meta-modeling of TNF-induced Tumor Resistance In Vitro

Fast and Accurate LSTM Meta-modeling of TNF-induced Tumor Re...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Marco P. Abrate Riccardo Smeriglio Roberta Bardini Alessandro Savino Stefano Di Carlo Dept. of Cell and Developmental Biology University College London London UK Dept. of Control and Computer Engineering Politecnico di Torino Turin Italy

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

Multi-level, hybrid models and simulations, among other methods, are essential to enable predictions and hypothesis generation in systems biology research. However, the computational complexity of these models poses a bottleneck, limiting the applicability of methodologies relying on large number of simulations, such as the Optimization via Simulation (OvS) of complex biological processes. Meta-models based on approximate surrogate models simplify multi-level simulations, maintaining accuracy while reducing computational costs. Among Artificial Neural Networks (ANNs), Long Short-Term Memory (LSTM) networks are well suited to handle sequential data, which often characterizes biological simulations. This paper presents an LSTM-based surrogate modeling approach for multi-level simulations of complex biological processes. Validation relies on the simulation of Tumor Necrosis Factor (TNF) administration to a 3T3 mouse fibroblasts tumor spheroid based on PhysiBoSS 2.0, a hybrid agent-based multi-level modeling framework. Results show that the proposed LSTM meta-model is accurate and fast compared with the simulator. In fact, it infers simulated behavior with an average relative error of 7.5%. Moreover, it is at least five orders of magnitude faster. Even considering the cost of training, this approach provides a faster, more accurate, and reusable surrogate of multi-scale simulations in computationally complex tasks, such as model-based OvS of biological processes.

关键词： Training Accuracy Biological system modeling Computational modeling Biological processes Metamodeling Predictive models Data models Long short term memory Tumors

来源：评论

学校读者我要写书评

暂无评论

DORAC: Dynamic-Objects Removal and Clustering for Object Detection

DORAC: Dynamic-Objects Removal and Clustering for Object Det...

引用

Cyber Science and Technology Congress (CyberSciTech), IEEE

作者： Kwanho Kiml Dokyung Kim Kanghyun Jo Dept. of Electrical Electronic and Computer Engineering (University of Ulsan) Intelligent Systems Laboratory Ulsan Republic of Korea

ISBN: (数字)9798331532093

ISBN: (纸本)9798331532109

Object detection using deep learning approach is being utilized for many fields such as autonomous driving, traffic analysis, etc. However, deep learning approach shows some fundamental limitations related to collecting datasets or explainabilty caused by its unclearness of the process for reaching a specific conclusion. This paper proposes DORAC which refers to Dynamic-Objects Removal And Clustering for object detection and does not use any deep neural networks. DORAC can be operated at videos captured by fixed camera. Frames of a video are averaged for removing dynamic objects. Then, the confidence map obtained from subtraction between video frame and averaged image is processed by thresholding and clustering. Experiments show that proposed method outperforms the state-of-the-art deep learning models trained with remarkable dataset by over 5% for our industry videos. In addition to metrics related to accuracy, frame rate of the method proposed in this paper is also 0.5 frames faster than YOLOv9-tiny model in CPU, averagely.

关键词： Deep learning Measurement Industries Accuracy Object detection Artificial neural networks Cameras Autonomous vehicles Videos

来源：评论

学校读者我要写书评

暂无评论

A Tensor Low-Rank Approximation for Value Functions in Multi-Task Reinforcement Learning

arXiv

引用

arXiv 2025年

作者： Rozada, Sergio Paternain, Santiago Bazerque, Juan Andrés Marques, Antonio G. Dept. of Signal Theory and Communications King Juan Carlos University Madrid Spain Dept. of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute TroyNY United States Dept. of Electrical and Computer Engineering University of Pittsburgh PittsburghPA United States

In pursuit of reinforcement learning systems that could train in physical environments, we investigate multi-task approaches as a means to alleviate the need for massive data acquisition. In a tabular scenario where the Q-functions are collected across tasks, we model our learning problem as optimizing a higher order tensor structure. Recognizing that close-related tasks may require similar actions, our proposed method imposes a low-rank condition on this aggregated Q-tensor. The rationale behind this approach to multi-task learning is that the low-rank structure enforces the notion of similarity, without the need to explicitly prescribe which tasks are similar, but inferring this information from a reduced amount of data simultaneously with the stochastic optimization of the Q-tensor. The efficiency of our low-rank tensor approach to multi-task learning is demonstrated in two numerical experiments, first in a benchmark environment formed by a collection of inverted pendulums, and then into a practical scenario involving multiple wireless communication devices. Copyright © 2025, The Authors. All rights reserved.

关键词： Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

A 0.77-pJ/bit 40-Gb/s/pin Single-Ended Hybrid DAC-Based Transmitter for Memory Interfaces

A 0.77-pJ/bit 40-Gb/s/pin Single-Ended Hybrid DAC-Based Tran...

引用

IEEE International Symposium on Circuits and systems (ISCAS)

作者： Sanghyuk Seo Yong-un Jeong Jeakwang Yun Jaewook Kim Suhwan Kim Dept. of Electrical and Computer Engineering Seoul National University Seoul Korea Dept. of Semiconductor Systems Engineering Sejong University Seoul Korea Dept. of DRAM Design SK Hynix Icheon Korea

ISBN: (数字)9798350330991

ISBN: (纸本)9798350331004

This paper presents a single-ended hybrid digital-to-analog converter (DAC)-based transmitter (TX) using four-level pulse-amplitude modulation (PAM-4) for memory interfaces. The proposed hybrid DAC composed of voltage mode (VM) driver unit segments and current mode (CM) driver unit segments reduces the total number of unit segments without increasing complexity in impedance matching. Therefore, the hybrid DAC significantly reduces area and the associated power consumption. Digital signal processing (DSP)-based feed-forward equalization (FFE) provides precise and flexible equalization. The proposed linearity enhancement technique compensates for additional non-linearity caused by short-length devices in the CM driver segments. The proposed TX operating at 40Gb/s/pin consumes 30.9mW with 1.0/0.9/0.15V supply voltages, achieving 0.77pJ/bit energy efficiency and a level separation mismatch ratio (RLM) of 0.989. The TX occupies an active area of 0.023mm 2 .

关键词： Power demand Transmitters Digital-analog conversion Impedance matching Memory management Linearity Voltage

来源：评论

学校读者我要写书评

暂无评论

SIMHEURISTIC AND LEARNHEURISTIC FOR SOLVING STOCHASTIC AND/OR DYNAMIC PORTFOLIO OPTIMIZATION PROBLEMS 11

SIMHEURISTIC AND LEARNHEURISTIC FOR SOLVING STOCHASTIC AND/O...

引用

11th Operational Research Society Simulation Workshop, SW 2023

作者： Li, Yuda Polat, Onur Juan, Angel A. Calvet, Laura Kizys, Renatas Dept. of Computer Science Universitat Oberta de Catalunya Barcelona Spain Dept. of Applied Statistics and Operations Research Universitat Politcnica de Valncia Alcoy Spain Dept. of Public Finance Bilecik Şeyh Edebali University Bilecik Turkey Dept. of Telecomm. and Systems Engineering Autonomous University of Barcelona Sabadell Spain Dept. of Banking and Finance Southampton Business School University of Southampton Southampton United Kingdom

ISBN: (纸本)9781713870951

Constructing portfolio by proper asset selection to maximize return and minimize risk has been considered an essential task for investment activities. Rich portfolio optimizations with realistic constraints are NP-hard problems and are commonly solved using metaheuristics. However, financial markets are characterized by their high volatility and uncertainty, and metaheuristics do not fully account for these random and/or dynamic components, which renders them unrealistic in the presence of heightened uncertainty and dynamism in financial markets. Therefore, this paper proposes a simulationoptimization approach specifically, a simheuristic algorithm to deal with the stochastic version of the problem and a learnheuristic algorithm for solving the dynamic version of the problem. Computational experiments are performed on a benchmark instance to illustrate the advantages of the proposed methodologies and analyze how the solutions change in response to a different degree of stochasticity, dynamism, and minimum required return. © SW *** rights reserved

关键词： Financial markets

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：