检索结果-内蒙古大学图书馆

IEEE/RAS-EMBS International Conference on Biomedical robotics and Biomechatronics (BioRob)

作者： Jhon P.F. Charaja Isabell Wochner Pierre Schumacher Winfried Ilg Martin Giese Christophe Maufroy Andreas Bulling Syn Schmitt Georg Martius Daniel F.B. Haeufle Hertie Institute for Clinical Brain Research and Centre for Integrative Neuroscience University of Tübingen Germany Institute of Computer Engineering (ZITI) Heidelberg University Germany Max Planck Institute for Intelligent Systems Tübingen Germany Institut for Industrial Manufacturing and Management (IFF) University of Stuttgart Germany Fraunhofer Institute for Manufacturing Engineering and Automation (IPA) Germany Institute for Visualization and Interactive Systems University of Stuttgart Germany Institute for Modelling and Simulation of Biomechanical Systems University of Stuttgart Germany Department of Computer Science University of Tübingen Germany

ISBN: (数字)9798350386523

ISBN: (纸本)9798350386530

Mimicking of human-like arm movement characteristics involves considering three factors during control policy synthesis: (a) task requirements, (b) noise during movement execution, and (c) optimality principles. Previous studies showed that when these factors (a-c) are considered individually, it is possible to synthesize arm movements that either kinematically match experimental data or reproduce the stereotypical triphasic muscle activation pattern. However, no quantitative comparison has assessed the realism of arm movements generated by each factor, nor has it been determined whether combining these factors results in movements with human-like kinematic characteristics and the triphasic muscle pattern. To investigate this, we used reinforcement learning to learn a control policy for a musculoskeletal arm model, aiming to discern which combination of factors (a-c) results in realistic arm movements according to four frequently reported stereotypical characteristics. Our findings indicate that incorporating velocity and acceleration requirements into the reaching task, employing reward terms that minimize mechanical work, hand jerk, and control effort, along with the inclusion of noise during movement, leads to realistic human arm movements by reinforcement learning. We expect that the gained insights will help in the future to better predict desired arm movements and corrective forces in wearable assistive devices.

关键词： Noise Reinforcement learning Kinematics Biomechatronics Muscles Manipulators Assistive devices Pattern matching

来源：评论

学校读者我要写书评

暂无评论

Development and Analysis of a Versatile Dataset of Speech, Real and Synthesized, of Arabic Learners

Development and Analysis of a Versatile Dataset of Speech, R...

引用

Computing and Information Technology (ICCIT-1441), 2020 International Conference on

作者： Mansour Alsulaiman Mohammed Algabri Hassan Mathkour Mohamed A. Bencherif Ghulam Muhammad Saad Al-Gahtani Mohammed Faisal Mohamed Amine Mekhtiche Taha Alfakih Computer Engineering Department College of Computer and Information Sciences King Saud University Riyadh Saudi Arabia Computer Science and Information Systems Department College of Applied Sciences AlMaarefa University Riyadh Saudi Arabia Computer Science Department College of Computer and Information Sciences King Saud University Riyadh Saudi Arabia Centre of Smart Robotics Research (CS2R) King Saud University Riyadh Saudi Arabia Arabic Language Institute King Saud University Saudi Arabia Center of AI & Robotics Kuwait College of Science and Technology (KCST) Kuwait

computer-Aided Pronunciation Training (CAPT) systems are gaining popularity recently due to the advancements in deep neural networks (DNN) and machine learning and the availability of databases of speech of language learners. Unfortunately, research in Arabic CAPT systems suffer from the lack of CAPT datasets compared to other languages. In this paper, we present the details of and the ideas used in the development of a versatile dataset of speech, real and synthesized, of Arabic Learners. To develop the dataset, we utilized an existing Arabic speech corpus, King Saud University Speech Database (KSU-SD). KSU-SD's main application was a speaker recognition system, but it was designed to be useful in other applications such as CAPT systems. KSU-SD includes a large number of speakers from diverse nationalities (Saudis, Arabs, and Non-Arabs). KSU-SD contains the recording of about 60 non-native speakers from more than 20 nationalities, hence we selected it to build the non-native Arabic-CAPT corpus. The developed corpus consists of transcribed, segmented, and annotated speech, which makes it suitable for building Arabic CAPT systems. In addition to presenting the details of developing the dataset, we also present an analysis of the text and speech errors of the dataset. The dataset was verified by many CAPT systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

BYTECounter: Improving Vehicle Turning-Movement Counting

BYTECounter: Improving Vehicle Turning-Movement Counting

引用

International Conference on intelligent Transportation

作者： Mateus Líbano Monteiro Alexandre Bernardino H. Sofia Pinto Computer Science: Artificial Intelligence and Cybersecurity from Instituto Superior Técnico Universidade de Lisboa Portugal Department of Electrical and Computer Engineering Institute for and Systems and Robotics of Instituto Superior Técnico Universidade de Lisboa Portugal INESC-ID and Department of Computer Science and Artificial Intelligence Instituto Superior Técnico Universidade de Lisboa Portugal

This paper proposes a methodology to enhance the accuracy of vehicle turning-movement counting in video footage. The methodology uses a three-component system comprising object detection, object tracking, and object turning-movement counting. To improve vehicle detection, a custom vehicle neural network called YOLOv7-vehicles-0.1 was trained using a unique vehicle dataset comprising COCO, KITTI, and OpenImages datasets. Additionally, a novel tracker named BYTECounter was developed, using two virtual drawing objects, Regions and Lines, to count vehicles' turning movements based on trajectory information. The performance of our system was evaluated using the UA-DETRAC dataset, achieving an overall Movement Counting Accuracy (MCAcc) of 93.5% with a processing rate of 82.86 frames per second. This paper demonstrates that BYTECounter outperforms the state-of-the-art tracker, BYTE, and highlights the superiority, in terms of counting metrics, of YOLOv7-vehicles-0.1 over YOLOv7 models trained on the COCO dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robotic Control Optimization Through Kernel Selection in Safe Bayesian Optimization

Robotic Control Optimization Through Kernel Selection in Saf...

引用

IEEE International Conference on robotics and Biomimetics

作者： Lihao Zheng Hongxuan Wang Xiaocong Li Jun Ma Prahlad Vadakkepat School of Data Science The Chinese University of Hong Kong Research Intern at the Singapore Institute of Manufacturing Technology A*STAR Shenzhen China Department of Electrical and Computer Engineering National University of Singapore Singapore Singapore Institute of Manufacturing Technology Agency for Science Technology and Research (A*STAR) Singapore Robotics and Autonomous Systems Thrust The Hong Kong University of Science and Technology (Guangzhou) Guangzhou China

ISBN: (数字)9798331509644

ISBN: (纸本)9798331509651

Control system optimization has long been a fundamental challenge in robotics. While recent advancements have led to the development of control algorithms that leverage learning-based approaches, such as SafeOpt, to optimize single feedback controllers, scaling these methods to high-dimensional complex systems with multiple controllers remains an open problem. In this paper, we propose a novel learning-based control optimization method, which enhances the additive Gaussian process-based Safe Bayesian Optimization algorithm to efficiently tackle high-dimensional problems through kernel selection. We use PID controller optimization in drones as a representative example and test the method on Safe Control Gym, a benchmark designed for evaluating safe control techniques. We show that the proposed method provides a more efficient and optimal solution for high-dimensional control optimization problems, demonstrating significant improvements over existing techniques.

关键词： Additives Trajectory tracking Optimization methods Control systems Bayes methods Kernel Robots Tuning Standards Quadrotors

来源：评论

学校读者我要写书评

暂无评论

iMTSP: Solving Min-Max Multiple Traveling Salesman Problem with Imperative Learning

arXiv

引用

arXiv 2024年

作者： Guo, Yifan Ren, Zhongqiang Wang, Chen The Flight Dynamics & Control/Hybrid Systems Lab Purdue University West LafayetteIN47907 United States Robotics Institute Carnegie Mellon University PittsburghPA15213 United States Lab Department of Computer Science and Engineering University at Buffalo NY14260 United States

This paper considers a Min-Max Multiple Traveling Salesman Problem (MTSP), where the goal is to find a set of tours, one for each agent, to collectively visit all the cities while minimizing the length of the longest tour. Though MTSP has been widely studied, obtaining near-optimal solutions for large-scale problems is still challenging due to its NP-hardness. Recent efforts in data-driven methods face challenges of the need for hard-to-obtain supervision and issues with high variance in gradient estimations, leading to slow convergence and highly sub-optimal solutions. We address these issues by reformulating MTSP as a bilevel optimization problem, using the concept of imperative learning (IL). This involves introducing an allocation network that decomposes the MTSP into multiple single-agent traveling salesman problems (TSPs). The longest tour from these TSP solutions is then used to self-supervise the allocation network, resulting in a new self-supervised, bilevel, end-to-end learning framework, which we refer to as imperative MTSP (iMTSP). Additionally, to tackle the high-variance gradient issues during the optimization, we introduce a control variate-based gradient estimation algorithm. Our experiments showed that these innovative designs enable our gradient estimator to converge 20× faster than the advanced reinforcement learning baseline, and find up to 80% shorter tour length compared with Google OR-Tools MTSP solver, especially in large-scale problems (e.g. 1000 cities and 15 agents). © 2024, CC BY.

关键词： Traveling salesman problem

来源：评论

学校读者我要写书评

暂无评论

Integrated Sensing and Communication System Based on Radio Frequency Resonance Beam

arXiv

引用

arXiv 2025年

作者： Guo, Yixuan Xia, Shuaifan Xiong, Mingliang Liu, Qingwen Zhou, Shengli Fang, Wen Jiang, Qingwei Yan, Gang Mu, Jiangchuan The Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai201210 China The College of Electronics and Information Engineering Tongji University Shanghai201804 China The Department of Computer Science and Technology Tongji University Shanghai201804 China The Department of Electrical and Computer Engineering University of Connecticut StorrsCT06269 United States The School of Physics Science and Engineering Tongji University Shanghai200092 China

To address the challenge of complex beam control in traditional multiple-input multiple-output (MIMO) systems, research has proposed establishing adaptive beam alignment by utilizing retro-directive antenna (RDA) arrays to create echo resonance between the base station (BS) and user equipment (UE), thereby reducing system computational load. However, in conventional resonant beam systems (RBS), the use of the same frequency for uplink and downlink inevitably leads to echo interference issues. Therefore, this paper proposes an innovative design for an resonance beam-based integrated sensing and communication (RB-ISAC) system to achieve efficient passive sensing and bidirectional communication. In this system, the UE does not actively transmit signals but instead relies on a passive phase conjugation and frequency conversion structure to separate the uplink and downlink carrier frequencies. Additionally, through effective compensation for signal propagation loss, resonance is achieved after multiple iterations, at which point the beam’s field distribution becomes a low-diffraction-loss, high-focusing pattern, automatically aligning the transmitter with the receiver. This enables high-precision passive positioning while facilitating uplink and downlink communication. Simulation results demonstrate that the proposed system can achieve resonance after multiple iterations, and can support uplink and downlink communication within a range of 5 meters while achieving passive direction of arrival (DOA) estimation with an error of less than 2 degrees. Copyright © 2025, The Authors. All rights reserved.

关键词： Directive antennas

来源：评论

学校读者我要写书评

暂无评论

Organic Heterojunction Enabled Near 1% H2o2 Photoproduction with Ultrafast Charge Separation

SSRN

引用

SSRN 2024年

作者： Liu, Jingjing Hu, Qiushi Liu, Shang Lin, Meng Zhang, Xiaosa Xie, Ying Chen, Xihan Shenzhen Key Laboratory of Intelligent Robotics and Flexible Manufacturing Systems SUSTech Energy Institute for Carbon Neutrality Department of Mechanical and Energy Engineering Southern University of Science and Technology Guangdong Shenzhen518055 China Key Laboratory for Rare Earth Chemistry and Application of Liaoning Province College of Science Shenyang University of Chemical Technology Liaoning Shenyang China

Hydrogen peroxide (H2O2) finds wide application in disinfection, chemical production, and bleaching owing to its sustainability, stability, and ease of storage compared to hydrogen. However, the current industrial engineering practices for producing photocatalysts used in H2O2 generation face challenges such as low yields with organic catalysts and high costs associated with inorganic catalysts. In light of these challenges, we investigate a viable method utilizing a simple condensation reflux technique to synthesize pure organic heterojunction photocatalysts featuring a π-π stacked structure derived from graphitized carbon nitride (melem) and meso-tetrakis (4-carboxyphenyl) porphyrin. This approach enables efficient H2O2 production from water, seawater, and oxygen without the necessity of sacrificial agents. In this study, the synthesized catalyst demonstrates an impressive H2O2 production rate of 106 mmol/h.g.L under seawater conditions and 77.67 mmol/h.g.L under pure water conditions in ambient air. Following a reaction time of 4.5 hours, the H2O2 concentration reaches 0.88 wt% with a volume of 5.0 mL. Transient absorption spectroscopy validates the formation of a heterojunction between melem and porphyrin. Notably, this heterogeneous catalyst efficiently facilitates water oxidation and oxygen reduction while suppressing both H2O2 decomposition and subsequent photodecomposition reactions, thereby addressing the challenges associated with organic catalyst yield. © 2024, The Authors. All rights reserved.

关键词： Absorption spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Directed Acyclic Transformer Pre-training for High-quality Non-autoregressive Text Generation

arXiv

引用

arXiv 2023年

作者： Huang, Fei Ke, Pei Huang, Minlie The CoAI group Tsinghua University Beijing China Institute for Artificial Intelligence State Key Lab of Intelligent Technology and Systems Beijing National Research Center for Information Science and Technology Department of Computer Science and Technology Tsinghua University Beijing China

Non-AutoRegressive (NAR) text generation models have drawn much attention because of their significantly faster decoding speed and good generation quality in machine translation. However, in a wider range of text generation tasks, existing NAR models lack proper pre-training, making them still far behind the pre-trained autoregressive models. In this paper, we propose Pre-trained Directed Acyclic Transformer (PreDAT) and a novel pre-training task to promote prediction consistency in NAR generation. Experiments on five text generation tasks show that our PreDAT remarkably outperforms existing pre-trained NAR models (+4.2 scores on average) and even achieves better results than pre-trained autoregressive baselines in n-gram-based metrics, along with 17 times speedup in throughput. Further analysis shows that PreDAT benefits from the unbiased prediction order that alleviates the error accumulation problem in autoregressive generation, which provides new insights into the advantages of NAR generation. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine translation

来源：评论

学校读者我要写书评

暂无评论

Optimizing Multi-DNN Inference on Mobile Devices through Heterogeneous Processor Co-Execution

arXiv

引用

arXiv 2025年

作者： Gao, Yunquan Zhang, Zhiguo Donta, Praveen Kumar Dehury, Chinmaya Kumar Wang, Xiujun Niyato, Dusit Zhang, Qiyang School of Computer Science and Technology Anhui Engineering Research Center for Intelligent Applications and Security of Industrial Internet Anhui University of Technology Anhui Ma’anshan243032 China Computer Science and Technology Anhui University of Technology Anhui Ma’anshan243032 China Department of Computer and Systems Sciences Stockholm University Stockholm106 91 Sweden Institute of Computer Science University of Tartu IISER Berhampur India College of Computing and Data Science Nanyang Technological University Singapore639798 Singapore Computer Science School Peking University Beijing100876 China

Deep Neural Networks (DNNs) are increasingly deployed across diverse industries, driving a growing demand to enable their capabilities on mobile devices. However, existing mobile inference frameworks are often rely on a single processor to handle each model’s inference, limiting hardware utilization and leading to suboptimal performance and energy efficiency. Expanding DNNs accessibility on mobile platforms requires more adaptive and resource-efficient solutions to meet increasing computational demands without compromising device functionality. Nevertheless, parallel inference of multiple DNNs on heterogeneous processors remains a significant challenge. Several works have explored partitioning DNN operations into subgraphs to enable parallel execution across heterogeneous processors. However, these approaches typically generate excessive subgraphs based solely on hardware compatibility, increasing scheduling complexity and memory management overhead. To address these limitations, we propose an Advanced Multi-DNN Model Scheduling (ADMS) strategy that optimizes multi-DNN inference across heterogeneous processors on mobile devices. ADMS constructs an optimal subgraph partitioning strategy offline, considering both hardware support of operations and scheduling granularity, while employing a processor-state-aware scheduling algorithm that dynamically balances workloads based on real-time operational conditions. This ensures efficient workload distribution and maximizes the utilization of available processors. Experimental results show that, compared to vanilla inference frameworks, ADMS reduced multi-DNN inference latency by 4.04×.MSC Codes 68T07, 68W40 Copyright © 2025, The Authors. All rights reserved.

关键词： Parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Multi-Objective Archiving

arXiv

引用

arXiv 2023年

作者： Li, Miqing López-Ibáñez, Manuel Yao, Xin The School of Computer Science University of Birmingham United Kingdom Alliance Manchester Business School University of Manchester United Kingdom The Research Institute of Trustworthy Autonomous Systems Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China

Most multi-objective optimisation algorithms maintain an archive explicitly or implicitly during their search. Such an archive can be solely used to store high-quality solutions presented to the decision maker, but in many cases may participate in the search process (e.g., as the population in evolutionary computation). Over the last two decades, archiving, the process of comparing new solutions with previous ones and deciding how to update the archive/population, stands as an important issue in evolutionary multi-objective optimisation (EMO). This is evidenced by constant efforts from the community on developing various effective archiving methods, ranging from conventional Pareto-based methods to more recent indicator-based and decomposition-based ones. However, the focus of these efforts is on empirical performance comparison in terms of specific quality indicators;there is lack of systematic study of archiving methods from a general theoretical perspective. In this paper, we attempt to conduct a systematic overview of multiobjective archiving, in the hope of paving the way to understand archiving algorithms from a holistic perspective of theory and practice, and more importantly providing a guidance on how to design theoretically desirable and practically useful archiving algorithms. In doing so, we also present that archiving algorithms based on weakly Pareto compliant indicators (e.g., ϵ-indicator), as long as designed properly, can achieve the same theoretical desirables as archivers based on Pareto compliant indicators (e.g., hypervolume indicator). Such desirables include the property limit-optimal, the limit form of the possible optimal property that a bounded archiving algorithm can have with respect to the most general form of superiority between solution sets. Copyright © 2023, The Authors. All rights reserved.

关键词： Multiobjective optimization

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：