检索结果-内蒙古大学图书馆

Acoustics, Speech, and Signal Processing Workshops (ICASSPW), IEEE International Conference on

作者： Wenjie Li Lei Chen Bing Lv Xiaobo Li Weijun Wang Kun Xu Xilun Ding School of Mechanical Engineering and Automation Beihang University Beijing BJ P.R.China Beijing Microelectronics Technology Institute Beijing BJ P.R.China China Academy of Aerospace Electronics Technology Beijing BJ P.R.China Robotics Engineering Center The 21ST Research Institute of China Electronic Technology Group Corporation Shanghai China

ISBN: (数字)9798350374513

ISBN: (纸本)9798350374520

Brain Computer Interfaces establishes an interaction mode between the human brain and external devices, and has broad application prospects. With the development of neuroscience and artificial intelligence technology, brain computer interfaces have rapidly evolved from a single perception stage to a multimodal cognition and complex interaction control stage. This article elaborates on the functional zoning of the brain and cerebellum and the latest research on brain signal classification. Afterwards, the latest research progress of deep learning based EEG signal processing technology was analyzed in depth from three key directions: brain network analysis, research on neurological diseases, cognitive analysis, and emotion recognition. This paper provides a detailed description of the typical structure and main functional unit design ideas of Brain Computer Interface chips, and analyzes the core and difficult technologies from the aspects of acquisition, 3D heterogeneous integration, low-power design, software and brain science research. Finally, based on typical Brain Computer Interface design and application cases, the future development trends of Brain Computer Interface chips and deep learning EEG technology are proposed.

关键词： Deep learning Neurological diseases Three-dimensional displays Neuroscience Pattern classification Signal processing Brain-computer interfaces

来源：评论

学校读者我要写书评

暂无评论

Dynamically Adaptive Cascading Updates for Hierarchical Digital Twin

SSRN

引用

SSRN 2024年

作者： Guo, Haoyu Wang, Shaoping Shi, Jian Lizzio, Fausto Guglieri, Giorgio School of Automation Science and Electrical Engineering Beihang University Beijing100191 China Department of Mechanical and Aerospace Engineering Polytechnic University of Turin Turin10129 Italy Beihang Ningbo Research Institute Ningbo315800 China Tianmushan Laboratory Hangzhou310023 China

Digital twin (DT) has emerged as a focal point of current research due to its advanced capability to dynamically map the evolution of physical equipment. The dynamic updating feature stands as one of DT’s core technological strengths. However, existing methods for updates often face challenges in efficiency and accuracy when dealing with complex model update tasks. To address the above fundamental challenges, this paper introduces a hierarchical DTs cascading update strategy to coordinate complex update tasks, thereby enhancing the efficiency of updates. It also provides the dynamically adaptive update algorithm for error threshold and batch size to optimize the update process based on signal change rates which helps to enhance the fidelity and real-time performance of the DT system. Finally, the paper validates the effectiveness of the proposed updating method by using oil leakage from the barrel-port plate of a piston pump as an illustrative example. The results confirm the significant improvement of this updating method in enhancing DT accuracy and updating efficiency. © 2024, The Authors. All rights reserved.

关键词： Leakage (fluid)

来源：评论

学校读者我要写书评

暂无评论

Distributed multi-agent target search and tracking with Gaussian process and reinforcement learning

arXiv

引用

arXiv 2023年

作者： Kim, Jigang Jang, Dohyun Jin Kim, H. Department of Mechanical and Aerospace Engineering Automation and Systems Research Institute Seoul National University 1 Gwanak-ro Gwanak-gu Seoul08826 Korea Republic of Department of Aerospace Engineering Automation and Systems Research Institute Seoul National University 1 Gwanak-ro Gwanak-gu Seoul08826 Korea Republic of Department of Aerospace Engineering Seoul National University 1 Gwanak-ro Gwanak-gu Seoul08826 Korea Republic of

Deploying multiple robots for target search and tracking has many practical applications, yet the challenge of planning over unknown or partially known targets remains difficult to address. With recent advances in deep learning, intelligent control techniques such as reinforcement learning have enabled agents to learn autonomously from environment interactions with little to no prior knowledge. Such methods can address the exploration-exploitation tradeoff of planning over unknown targets in a data-driven manner, eliminating the reliance on heuristics typical of traditional approaches and streamlining the decision-making pipeline with end-to-end training. In this paper, we propose a multi-agent reinforcement learning technique with target map building based on distributed Gaussian process. We leverage the distributed Gaussian process to encode belief over the target locations and efficiently plan over unknown targets. We evaluate the performance and transferability of the trained policy in simulation and demonstrate the method on a swarm of micro unmanned aerial vehicles with hardware experiments. © 2023, CC BY-NC-ND.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Online optimization over Riemannian manifolds

The Journal of Machine Learning Research

引用

The Journal of Machine Learning research 2023年第1期24卷 3762-3828页

作者： Xi Wang Zhipeng Tu Yiguang Hong Yingyi Wu Guodong Shi Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing P. R. China and Australian Center for Robotics School of Aerospace Mechanical and Mechatronic Engineering The University of Sydney NSW Australia Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai P. R. China Department of Mathematics University of Chinese Academy of Sciences Beijing P. R. China Australian Center for Robotics School of Aerospace Mechanical and Mechatronic Engineering The University of Sydney NSW Australia

Online optimization has witnessed a massive surge of research attention in recent years. In this paper, we propose online gradient descent and online bandit algorithms over Riemannian manifolds in full information and bandit feedback settings respectively, for both geodesically convex and strongly geodesically convex functions. We establish a series of upper bounds on the regrets for the proposed algorithms over Hadamard manifolds. We also find a universal lower bound for achievable regret on Hadamard manifolds. Our analysis shows how time horizon, dimension, and sectional curvature bounds have impact on the regret bounds. When the manifold permits positive sectional curvature, we prove similar regret bound can be established by handling non-constrictive project maps. In addition, numerical studies on problems defined on symmetric positive definite matrix manifold, hyperbolic spaces, and Grassmann manifolds are provided to validate our theoretical findings, using synthetic and real-world data.

关键词： online optimization Riemannian manifolds Riemannian optimization gradient estimation Fréchet mean

来源：评论

学校读者我要写书评

暂无评论

Investigation on the multi-solution problem of the kinetostatics of cable-driven continuum manipulators

Investigation on the multi-solution problem of the kinetosta...

引用

IEEE International Conference on robotics and automation (ICRA)

作者： Yicheng Dai Zuan Li Xin Wang Han Yuan School of Mechanical Engineering and Automation Harbin Institute of Technology (Shenzhen) Shenzhen China Guangdong Provincial Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robots Key University Laboratory of Space Mechanisms and Robots of Guangdong Shenzhen China Aerospace Research and Development Center Shanghai Aerospace Control Technology Institute Shanghai China

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

Cable-driven continuum manipulators have gained considerable attention due to their high dexterity and inherent structural compliance, making them a popular research topic. However, previous studies have overlooked the kinetostatics of these manipulators, which can result in a multi-solution problem. This issue is critical as having multiple equilibrium states can lead to erroneous estimations of the manipulator's profile. To address this issue, the kinetostatic model is presented and simulations based on both the interval analysis method and the commonly used floating-point optimization algorithm are conducted under the same actuating forces and external loads. Results show that there are multiple solutions to the kinetostatics of cable-driven continuum manipulators with constant cross section or variable cross section. This paper fills a gap in the current literature and offers valuable insights for researchers in the field of cable-driven continuum manipulators.

关键词： Analytical models Three-dimensional displays Nonlinear equations Manipulators Mathematical models Filling Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Toward the smooth mesh climbing of a miniature robot using bioinspired soft and expandable claws

arXiv

引用

arXiv 2022年

作者： Wang, Hong Liu, Peng Ngoc, Phuoc Thanh Tran Li, Bing Li, Yao Sato, Hirotaka State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin150001 China The School of Mechanical Engineering and Automation Harbin Institute of Technology Shenzhen518055 China School of Mechanical and Aerospace Engineering Nanyang Technological University Singapore Singapore

While most micro-robots face difficulty traveling on rugged and uneven terrain, beetles can walk smoothly on the complex substrate without slipping or getting stuck on the surface due to their stiffness-variable tarsi and expandable hooks on the tip of tarsi. In this study, we found that beetles actively bent and expanded their claws regularly to crawl freely on mesh surfaces. Inspired by the crawling mechanism of the beetles, we designed an 8-cm miniature climbing robot equipping artificial claws to open and bend in the same cyclic manner as natural beetles. The robot can climb freely with a controllable gait on the mesh surface, steep incline of the angle of 60°, and even transition surface. To our best knowledge, this is the first micro-scale robot that can climb both the mesh surface and cliffy incline. © 2022, CC BY.

关键词： Robots

来源：评论

学校读者我要写书评

暂无评论

research on modular charging systems based on digital twins

Research on modular charging systems based on digital twins

引用

Power Electronics, Computer Applications (ICPECA), IEEE International Conference on

作者： Qicheng Dong Zhigang Xu Junyi Wang Xiao Yang Linlin Fan Wei Li College of Mechanical and Electrical Engineering Shenyang Aerospace University Shenyang China Shenyang Institute of Automation Chinese Academy of Sciences Institute of Robotics and Intelligent Manufacturing Innovation Chinese Academy of Sciences Shenyang China Northwestern Industries Group Co. Ltd Xi’an China

The development of ammunition loading technology plays a vital role in the improvement of my country's national defense construction level. Compared with the traditional charge, the modular charge has the advantages of high interchangeability and modularity, and can realize the rapid filling of the charge. However, for the application of modular charging technology, when building unit modules, it is still necessary to manually teach the charging target point, and the modular charging technology has the disadvantage of being difficult to adapt to flexibility. Therefore, in the face of the diversity of ammunition, modular charge is not conducive to the rapid and efficient production of ammunition. The current automation technology is difficult to improve the flexible adaptability of modular charge. Aiming at the disadvantages of modular charging technology, a modular flexible charging system based on digital twin is proposed, including the construction of virtual charging workstation, key point identification, automatic trajectory generation and automatic robot program generation. The system greatly improves the production efficiency of modular charge, provides a reference for the research of modular flexible charge, and improves the digital manufacturing level of ammunition production.

关键词： Drugs Robot kinematics Loading Production Power electronics Filling Digital twins

来源：评论

学校读者我要写书评

暂无评论

Rough Set Based Contour Approximation From Flickering Depth Images

Rough Set Based Contour Approximation From Flickering Depth ...

引用

Emerging Smart Computing and Informatics (ESCI), Conference on

作者： Soma Banerjee Robotics & Automation CSIR-Central Mechanical Engineering Research Institute Durgapur India

ISBN: (数字)9781728185194

ISBN: (纸本)9781728185200

This paper describes a Rough Set based method to approximate the contour of flickering depth images taken from coded light camera. Due to the flickering effect sometimes some contour pixels are appearing and disappearing at the same location which leads to problem in detecting the true contour and corners of the object. This effect may be more or less depending on calibration and selection of frames per second (FPS) factor. Even after everything is done this effect of flicker is present in depth camera images and it comes inevitably with the technology used for imaging. The concept of Rough Set is applied to determine contour with accuracy. The persistent pixels of the object image are kept in lower approximation and boundary pixels which are not consistent are kept in upper approximation of contour set. Pixels from both lower and upper approximation are chosen to delineate the contour of object in the image.

关键词： Three-dimensional displays Laser noise Rough sets Measurement by laser beam Cameras Calibration Informatics

来源：评论

学校读者我要写书评

暂无评论

MPC-Based Human-Accompanying Control Strategy for Improving the Motion Coordination Between the Target Person and the Robot

MPC-Based Human-Accompanying Control Strategy for Improving ...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Jianwei Peng Zhelin Liao Hanchen Yao Zefan Su Yadan Zeng Houde Dai Fujian Institute of Research on The Structure of Matter Chinese Academy of Sciences Fuzhou China Fujian College University of Chinese Academy of Sciences Jinjiang China Robotics Research Centre School of Mechanical and Aerospace Engineering Nanyang Technological University Singapore

Social robots have gained widespread attention for their potential to assist people in diverse domains, such as living assistance and logistics transportation. Human-accompanying, i.e., walking side-by-side with a person, is an expected and essential capability for social robots. However, due to the complexity of motion coordination between the target person and the mobile robot, the accompanying action is still unstable. In this study, we propose a human-accompanying control strategy to improve the motion coordination for better practicability of the human-accompanying robot. Our approach allows the robot to adapt to the motion variations of the target person and avoid obstacles while accompanying them. First, a human-robot interaction model based on the separation-bearing-orientation scheme is developed to ascertain the relative position and orientation between the robot and the target person. Then, a human-accompanying controller based on behavioral dynamics and model predictive control (MPC) is designed to avoid obstacles and simultaneously track the direction and velocity of the target person. Experimental results indicate that the proposed method can effectively achieve side-by-side accompanying by simultaneously controlling the relative position, direction, and velocity between the target person and robot.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SoftGPT: Learn Goal-Oriented Soft Object Manipulation Skills by Generative Pre-Trained Heterogeneous Graph Transformer

SoftGPT: Learn Goal-Oriented Soft Object Manipulation Skills...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Junjia Liu Zhihao Li Wanyu Lin Sylvain Calinon Kay Chen Tan Fei Chen Department of Mechanical and Automation Engineering T-Stone Robotics Institute The Chinese University of Hong Kong Hong Kong SAR Department of Computing The Hong Kong Polytechnic University Hong Kong SAR Idiap Research Institute Martigny Switzerland

Soft object manipulation tasks in domestic scenes pose a significant challenge for existing robotic skill learning techniques due to their complex dynamics and variable shape characteristics. Since learning new manipulation skills from human demonstration is an effective way for robot applications, developing prior knowledge of the representation and dynamics of soft objects is necessary. In this regard, we propose a pretrained soft object manipulation skill learning model, namely SoftGPT, that is trained using large amounts of exploration data, consisting of a three-dimensional heterogeneous graph representation and a GPT-based dynamics model. For each downstream task, a goal-oriented policy agent is trained to predict the subsequent actions, and SoftGPT generates the consequences of these actions. Integrating these two approaches establishes a thinking process in the robot's mind that provides rollout for facilitating policy learning. Our results demonstrate that leveraging prior knowledge through this thinking process can efficiently learn various soft object manipulation skills, with the potential for direct learning from human demonstrations.

关键词：

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：