检索结果-内蒙古大学图书馆

IEEE Transactions on Circuits and systems for Artificial Intelligence 2025年第2期2卷 114-125页

作者： Mikail Yayla Ming-Liang Wei Simon Thomann Albi Mema Chia-Lin Yang Jian-Jia Chen Hussam Amrouch Design Automation for Embedded Systems Group TU Dortmund University Dortmund Germany Lamarr Institute for Machine Learning and Artificial Intelligence Dortmund Germany Department of the Graduate Institute of Electronics Engineering National Taiwan University Taipei Taiwan Technical University of Munich (TUM) Munich Germany TUM School of Computation Information and Technology Chair of AI Processor Design Munich Institute of Robotics and Machine Intelligence Munich Germany Department of Computer Science and Information Engineering Graduate Institute of Electronics Engineering National Taiwan University Taipei Taiwan Department of Computer Science TU Dortmund University Dortmund Germany Lamarr Institute Dortmund Germany

Accelerating the immense workloads of Neural Networks (NNs) is a critical challenge, and analog computing-based approaches present a promising solution. Among these, the Integrate-and-Fire (IF) Spiking Neural Network (SNN) stands out for its potential in efficient neural computation. However, achieving high inference accuracy through precise multiply-accumulate (MAC) operations necessitates a large membrane capacitor—224 to 547 times larger than the synapse array at the 14nm technology node—resulting in prohibitive area costs. Additionally, the large capacitor size introduces higher energy consumption and longer access times. Thus, a key research challenge is maintaining accuracy while mitigating the costs associated with the large membrane capacitor. In this work, we propose a HW/SW Codesign method, called CapMin, for capacitor size minimization in analog computing IF-SNNs. CapMin minimizes the capacitor size by reducing the number of spike times needed for accurate operation of the HW, based on the absolute frequency of MAC level occurrences in the SW. To increase the operation of IF-SNNs to current variation, we propose the method CapMin-V, which trades capacitor size for protection based on the reduced capacitor size found in CapMin. In our experiments, CapMin achieves more than a 14 $\times$ reduction in capacitor size along with a 34% reduction in energy consumption over the state of the art, while CapMin-V achieves increased variation tolerance in the IF-SNN operation, requiring only a small increase in capacitor size.

关键词： Capacitors Circuits Accuracy Capacitance Micromechanical devices Costs Neurons Integrated circuit modeling Spiking neural networks Hardware

来源：评论

学校读者我要写书评

暂无评论

Design of a Novel Autonomous Docking System With a Supporting Robotic Fish

Design of a Novel Autonomous Docking System With a Supportin...

引用

2023 IEEE International Conference on Real-Time Computing and robotics, RCAR 2023

作者： Dong, Huijie Zhao, Chunyang Qiao, Tiezhu Wang, Jian Chen, Di Meng, Yan Yu, Junzhi Taiyuan University of Technology Key Laboratory of Advanced Transducers and Intelligent Control System Ministry of Education Taiyuan030024 China Chinese Academy of Sciences Laboratory of Cognitive and Decision Intelligence for Complex System Institute of Automation Beijing100190 China Peking University State Key Laboratory for Turbulence and Complex Systems Department of Advanced Manufacturing and Robotics College of Engineering Beijing100871 China Guangdong Ocean University College of Electronics and Information Engineering Zhanjiang524088 China

ISBN: (纸本)9798350327182

Underwater supporting robots serving as a relay of energy supplements and communication for other underwater equipment are promising for ocean exploration, development, and protection. This paper proposes a novel autonomous docking system centered on a designed supporting robotic fish named 'CourierFish'. Specifically, CourierFish is capable of docking with a surface dock station for supplying itself and docking with a seafloor platform for supporting equipment in the platform. A visual navigation scheme integrating LED and ArUco markers is presented for accurate localization. The control approach for docking motion is also illustrated. Simulations and aquatic experiments are performed to verify the feasibility of the proposed docking system. The obtained results lay a solid foundation for the construction of various underwater equipment and robot networks. © 2023 IEEE.

关键词： Fish

来源：评论

学校读者我要写书评

暂无评论

Uapgo: Uncertainty-Aware Pose Graph Optimization for Lidar-Based Slam systems

SSRN

引用

SSRN 2024年

作者： Zhi, Tian Xu, Jie Ma, Yongxin Yu, Wenlu Liu, Shenglan Yuan, Shenghai College of Mechanical and Electrical Engineering Nanjing University of Aeronautics and Astronautics Jiangsu Nanjing210016 China School of Electrical and Electronic Engineering Nanyang Technological University Singapore639798 Singapore State Key Laboratory of Robotics and Systems Harbin Institute of Technology Harbin150001 China School of Mechanical Engineering Shandong University Jinan250061 China

Addressing uncertainty errors, although crucial, may sometimes be deprioritized in the optimization process ofLiDAR-based SLAM systems due to the inherent complexity of the task. The practice of applying uniform weights, which simplifies the uncertainty management, when incorporating factors can inadvertently degrade the pose graph optimization, particularly within complex environments. In response to this, we propose anuncertainty-aware pose graph optimization technique, along with an innovative keyframe selection strategy. We develop a model for estimating the covariance of relative pose transformations, utilizing frontend odometry output and the Huber robust kernel function. This model assigns weights based on the uncertainty error of pose transformations. Upon detection of a loop closure, the system optimizes keyframe poses, effectively reducing cumulative error. Our keyframe selection strategy takes into account the maximum eigenvalue of the translational component of pose covariance as an uncertainty measure. If this value surpasses a certain threshold, the pose is selected as a keyframe, thereby enhancing data precision and computational efficiency, and subsequently improving the SLAM system’s ability to handle uncertainty. Ablation studies validate our approach, demonstrating its effectiveness in enhancing performance, especially in complex environments and conditions of data degradation. Wealso provide the associated open-source code for further research. © 2024, The Authors. All rights reserved.

关键词： Eigenvalues and eigenfunctions

来源：评论

学校读者我要写书评

暂无评论

Robotic dual-arm patch clamp for neuron excitability research under mechanical stimulation

Robotic dual-arm patch clamp for neuron excitability researc...

引用

第43届中国控制会议

作者： Biting Ma Chaoyu Cui Ruimin Li Minghui Li Jinyu Qiu Ke Li Mingzhu Sun Xin Zhao Qili Zhao National Key Laboratory of Intelligent Tracking and Forecasting fo Infectious Diseases Engineering Research Center of Trusted BehaviorIntelligence Ministry of Education Tianjin Key Laboratory of intelligentRobotic (tikLIR) Institute of Robotics and Automatic Information System(IRAlS) Nankai University Institute of Intelligence Technology and Robotic Systems Shenzhen Research Institute of Nankai University

ISBN: (数字)9789887581581

ISBN: (纸本)9798350366907

The dual-arm patch clamp technique, using one micropipette electrode to patch the target cell to record its electric signals with the other micropipette mechanically stimulating the target cell, is a significant technique in neuroscience for studying neuron excitability under mechanical stimulation. A key step for dual-arm patch clamp operation is to coordinate two micropipettes to reach the target cells in the narrow brain tissue environment. The traditional manual operations face issues such as inaccurate workspace estimation, clumsy movements, and collision risks of the two micropipettes, resulting in low operational efficiency. Addressing the above issues, a robotic dual-arm patch clamp method was developed for neuron excitability research under mechanical stimulation in this paper. First, the reachable workspace of the dual-arm was calculated based on a kinematics model. Then, a robotic dual arm patch clamp operation procedure is proposed to approach, whole-cell patch, and mechanically stimulate the target cell. The experiment results on neurons in mouse brain slices show that the success rate of dual micropipette contacting cell surface reaches up to 90%(18/20) at an average speed of 6 mins/cell, which are both significantly improved in comparison to manual operation. An increase in neuron excitability in the primary visual cortex(V1) was found under the mechanical stimulation, proving the feasibility of the proposed system for studying the impact of mechanical stimulation on ion channel activities.

关键词： Robotic Patch Clamp Robotic Cell Measurement Workspace Calculation Whole-cell Recording Mechanical Stimulation

来源：评论

学校读者我要写书评

暂无评论

Edge Detectors Can Make Deep Convolutional Neural Networks More Robust

arXiv

引用

arXiv 2024年

作者： Ding, Jin Zhao, Jie-Chao Sun, Yong-Zhi Tan, Ping Wang, Jia-Wei Ma, Ji-En Fang, You-Tong School of Automation and Electrical Engineering Key Institute of Robotics of Zhejiang Province Zhejiang University of Science and Technology Hangzhou310023 China State Key Laboratory of Fluid Power and Mechatronic Systems Zhejiang University Hangzhou310027 China School of Electrical Engineering Zhejiang University Hangzhou310027 China

Deep convolutional neural networks (DCNN for short) are vulnerable to examples with small perturbations. Improving DCNN’s robustness is of great significance to the safety-critical applications, such as autonomous driving and industry automation. Inspired by the principal way that human eyes recognize objects, i.e., largely relying on the shape features, this paper first employs the edge detectors as layer kernels and designs a binary edge feature branch (BEFB for short) to learn the binary edge features, which can be easily integrated into any popular backbone. The four edge detectors can learn the horizontal, vertical, positive diagonal, and negative diagonal edge features, respectively, and the branch is stacked by multiple Sobel layers (using edge detectors as kernels) and one threshold layer. The binary edge features learned by the branch, concatenated with the texture features learned by the backbone, are fed into the fully connected layers for classification. We integrate the proposed branch into VGG16 and ResNet34, respectively, and conduct experiments on multiple datasets. Experimental results demonstrate the BEFB is lightweight and has no side effects on training. And the accuracy of the BEFB integrated models is better than the original ones on all datasets when facing FGSM, PGD, and C&W attacks. Besides, BEFB integrated models equipped with the robustness enhancing techniques can achieve better classification accuracy compared to the original models. The work in this paper for the first time shows it is feasible to enhance the robustness of DCNNs through combining both shape-like features and texture features. © 2024, CC BY-NC-SA.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Artificial Neural Networks for the Forward Kinematics of a SCARA Manipulator: A Comparative Study with Two Datasets

Artificial Neural Networks for the Forward Kinematics of a S...

引用

Emerging Technologies for Sustainability and Intelligent systems (ICETSIS), ASU International Conference in

作者： Rania Bouzid Jyotindra Narayan Hassène Gritli RISC Laboratory (LR16ES07) Higher Institute of Information and Communication Technologies University of Carthage Borj Cedria Tunis Tunisia Department of Mechanical Engineering Mechatronics and Robotics Laboratory Indian Institute of Technology Guwahati Guwahati Assam India Laboratory of Robotics Informatics and Complex Systems (RISC Lab - LR16ES07) National Engineering School of Tunis University of Tunis El Manar Le Belvédère Tunis Tunisia

In the field of robotics and automation, the precise determination of manipulator positions is a fundamental aspect with widespread applications. Despite the significance of analytical methods, solving the Forward Kinematics (FK) for Selective Compliance Assembly Robot Arm (SCARA) manipulator robots can be complex and computationally intensive. Our research introduces a method to address the FK problem for the four degrees of freedom (4-DoF) SCARA manipulator through Artificial Neural Networks (ANNs). We utilize two different generated datasets, one with a fixed step size and another one based on a sinusoidal signal, for training the ANNs. This choice of datasets allows us to assess the ANN s' generalization capabilities across a range of operating conditions. Thereafter, the study is planned to investigate the ANN's performance while employing three distinct training algorithms: Levenberg-Marquardt (LM), Bayesian Regularization (BR), and Scaled Conjugate Gradient (SCG). Through a comprehensive comparison of the performance of different ANN models, different training algorithms, and the two adopted generated datasets, this study provides valuable insights into selecting the most optimal training configurations for the SCARA robot's FK solutions. We employ the Mean Squared Error (MSE) and the error histogram as two performance metrics to assess the accuracy of various ANNs. The results reveal that optimal MSE outcomes were obtained when utilizing a balanced architecture with three hidden layers in the analysis of the sinusoidal-signal-based datasets.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-Class Gait Phase Recognition using Machine Learning Models with Two Training Methods

Multi-Class Gait Phase Recognition using Machine Learning Mo...

引用

International Conference on Computing and Networking Technology (ICCNT)

作者： Amal Mekni Jyotindra Narayan Hassène Gritli RISC Laboratory (LR16ES07) Borj Cedria Higher Institute of Information and Communication Technologies University of Carthage Tunis Tunisia Department of Mechanical Engineering Mechtronics and Robotics Laboratory Indian Institute of Technology Guwahati Guwahati Assam India Laboratory of Robotics Informatics and Complex Systems RISC Lab - LR16ES07) National Engineering School of Tunis University of Tunis El Manar Le Belvédère Tunis Tunisia

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Walking is a fundamental human activity, yet understanding its intricacies is crucial for diagnosing and treating various gait abnormalities and musculoskeletal disorders. This study investigates the machine learning-based classification of gait phases into five subphases with two training methods. Using data from 100 individuals obtained from an open-source platform. The gait cycle was divided into the following subphases: Initial Contact, Loading Response, Mid Stance, Terminal Stance, and Pre-Swing. In the first approach, a stratified random sampling method was employed, allocating 80% of data from each subphase for training and 20% for testing. The second method involved training with data from 80% participants and testing with data from 20% participants. Out of different ML algorithms, the Random Forest (RF) achieves the highest performance (99.2%) with similar results across both methods. The k-Nearest Neighbors (k-NN), the Logistic Regression (LR), the Decision Tree (DT), and the Support Vector Machine (SVM) show comparable performance ($\sim 91 \%$) with marginal improvement with the second method. Naive Bayes (NB) has a significantly lower performance ($\sim 77.7 \%$). These findings suggest that RF is a robust and effective algorithm for this task. Furthermore, the study indicates that applying machine learning techniques to datasets divided by subjects presents promising opportunities for creating precise gait analysis systems. This can assist clinicians and researchers in enhancing patient care and gaining deeper insights into human locomotion.

关键词： Training Support vector machines Radio frequency Musculoskeletal system Logistic regression Machine learning algorithms Nearest neighbor methods Sampling methods Random forests Testing

来源：评论

学校读者我要写书评

暂无评论

Terrain Classification Based on Wheel-terrain Interaction Measurements using Aside-Wheel Camera

Terrain Classification Based on Wheel-terrain Interaction Me...

引用

IEEE International Conference on robotics and Biomimetics

作者： Feng Xue Longteng Hu Chen Yao Jin Wei Yunzhou Li Peichen Wang Zheng Zhu Zhenzhong Jia Department of Mechanical and Energy Engineering Shenzhen Key Laboratory of Biomimetic Robotics and Intelligent Systems Southern University of Science and Technology (SUSTech) Shenzhen China SUSTech Guangdong Provincial Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities Shenzhen China Harbin of Institute of Technology (HIT) Harbin China

The scientific applications over a decade have demonstrated the robust practicality of wheeled rovers, which are capable of undertaking exploration tasks in various unstructured terrains. However, sinkage and slippage are extremely common on sandy terrains. Fortunately, these risks can be mitigated in a timely manner through the perception of terrain types. Thus, terrain classification technology has become a key solution to addressing these issues. In this paper, we propose a new terrain classification pipeline for a rover over sandy terrains. It exploits the aside-wheel sensor suite to measure sinkage, contact angle, slip ratio, and normal force. Due to the strong correlation between contact characteristics and terrain types form our experiments, we can infer terrain types based on wheel-terrain interaction parameters. We train various machine learning models and obtained classification results, which is evaluated using the record dataset form a singlewheel testbed. Through Experimental comparison, most models achieve high accuracy (over 90%), we analyze the reasons behind performance discrepancies among different models and recommend LGBM as the preferred approach. In the end, we transfer the models to use data collected from rover field tests and achieve a higher level of accuracy, which proves the effectiveness of our approach.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Key Parameters Prediction for Intelligent Control and Energy Optimization Based on MIC-BWO-XGBoost 2

Key Parameters Prediction for Intelligent Control and Energy...

引用

2nd International Conference on Cyber-Energy systems and Intelligent Energy, ICCSIE 2024

作者： Lv, Changyao Wang, Yaqi Chen, Yudong Yu, Zhuo Hu, Guoliang Wu, Hongliang Li, Shuai Key Laboratory of Networked Control Systems Chinese Academy of Sciences Shenyang110016 China Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110169 China University of Chinese Academy of Sciences Beijing100049 China College of Information Engineering Shenyang University of Chemical Technology Shenyang110142 China Shanghai Aircraft Manufacturing Co. Ltd. Shanghai201324 China

ISBN: (纸本)9798350373318

In the aviation industry, anodic oxidation serves as a core process for surface modification of aluminum alloy components, albeit accompanied by high energy consumption. Accurate prediction of key parameters can offer decision support for the intelligent control and energy optimization of anodic oxidation. However, the limited sample size resulting from the high cost of data collection and the complex nonlinear patterns caused by the synergy of multiple factors pose a challenge in accurately predicting key parameters. Therefore, we propose a key parameters prediction method based on the Maximum Information Coefficient (MIC), Beluga Whale Optimization algorithm (BWO), and XGBoost algorithm. Considering the adverse impact of low correlation parameters on prediction accuracy, we first use MIC to extract features from the original data. Then, XGBoost is used to integrate multiple decision trees to ensure the strong fitting ability for nonlinear relationships. Compared to deep learning models, XGBoost performs well even in a limited sample size. In addition, considering XGBoost is prone to overfitting training data as an ensemble learning algorithm, we employ the BWO to optimize four key parameters of the model that are difficult to choose, thereby enhancing model's generalization. The model's performance is evaluated using Mean Squared Error (MSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE), with values of 0.04388, 0.09410, and 0.20730%, respectively. The experimental results demonstrate that the proposed method outperforms traditional approaches, enabling precise prediction of key parameters and holding significant value for the intelligent control and energy optimization of anode oxidation processes. © 2024 IEEE.

关键词： Anodic oxidation

来源：评论

学校读者我要写书评

暂无评论

An Augmented Reality Application for Upper-Limb Rehabilitation Supported by a Robotic Arm

An Augmented Reality Application for Upper-Limb Rehabilitati...

引用

Experiment@ International Conference (exp.at)

作者： José Carlos Rodrigues Maria Teresa Restivo Paulo Menezes LAETA-INEGI Faculty of Engineering University of Porto Porto Portugal LAETA-INEGI University of Porto Porto Portugal A3ES Lisbon Portugal Department of Electrical and Computer Engineering University of Coimbra Institute of Systems and Robotics Coimbra Portugal

ISBN: (数字)9798350314403

ISBN: (纸本)9798350314410

This work looks at the potential use of a robotic arm and augmented reality application, with a head-mounted display, in an upper-limb rehabilitation context. A sample application using a robotic simulator and a cross-sectional study using a real robotic arm are shown. Results from the usability evaluation are positive and encourage further development, but some identified problems must be corrected before further evaluations.

关键词： Head-mounted displays Sociology Medical treatment Collaborative robots Manipulators Real-time systems Usability

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：