检索结果-内蒙古大学图书馆

Design and Optimization of an Untethered High-Performance Robotic Tuna

IEEE-ASME TRANSACTIONS ON MECHATRONICS 2022年第5期27卷 4132-4142页

作者： Tong, Ru Wu, Zhengxing Chen, Di Wang, Jian Du, Sheng Tan, Min Yu, Junzhi Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China Peking Univ Coll Engn Dept Adv Mfg & Robot BIC ESATState Key Lab Turbulence & Complex Syst Beijing 100871 Peoples R China

For bioinspired underwater robots, it is a great challenge to achieve both high swimming speed and steering maneuverability. To this end, this article presents an untethered high-performance robotic tuna through mechanism optimization and steering strategies design. First, combining the advantages of single-joint and multijoint robotic fish, a novel mechanism of redundant joints is designed to enrich the swimming patterns of the robotic tuna. Next, a three-dimensional dynamic model is established, and the model parameters are accurately identified by experimental data. With the aid of the model and experiments, the speed performance is optimized under the key parameters of tail fin, such as size and flexibility. Further, from the perspective of imitating tuna and the demand of actual operations, two steering strategies are proposed to improve the steering performance. Finally, extensive simulation and experiments verify the effectiveness of the proposed methods. The obtained results reveal that the untethered robotic tuna can achieve both high swimming speed with 2.26 m/s (equivalent to 3.13 body lengths per second) and outstanding steering maneuverability with 0.48 body lengths in turning radius, providing valuable insight into performing special missions in cluttered ocean environments.

关键词： High maneuverability high swimming motion control motion optimization robotic tuna

来源：评论

学校读者我要写书评

暂无评论

Image Denoising of Seam Images With Deep Learning for Laser Vision Seam Tracking

引用

IEEE SENSORS JOURNAL 2022年第6期22卷 6098-6107页

作者： Yang, Lei Fan, Junfeng Huo, Benyan Li, En Liu, Yanhong Zhengzhou Univ Sch Elect Engn Zhengzhou 450001 Henan Peoples R China Robot Percept & Control Engn Lab Zhengzhou 450001 Henan Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China

Seam tracking with structured light vision has been widely applied into the robot welding. And the precise laser stripe extraction is the premise of automatic laser vision seam tracking. However, conventional laser stripe extraction methods based on image processing have the shortcomings of poor flexibility and robustness, which are easily affected by considerable image noises in the welding processing, such as arc light, smoke, and splash. To address this issue, inspired by image segmentation, with the strong contextual feature expression ability of deep convolutional neural network (DCNN), a novel image denoising method of seam images is proposed in this paper for automatic laser stripe extraction to serve intelligent robot welding applications, such as seam tracking, seam type detection, weld bead detection, etc. With the deep encoder-decoder network framework, aimed at the information loss issue by multiple convolutional and pooling operations in DCNNs, an attention dense convolutional block is proposed to extract and accumulate multi-scale feature maps. Meanwhile, a residual bi-directional ConvLSTM block (BiConvLSTM) is proposed to effectively learn multi-scale and long-range spatial contexts from local feature maps. Finally, a weighted loss function is proposed for model training to address the class unbalanced issue. Combined with the seam image set, the experimental results show that the proposed image denoising network could correctly extract the laser stripes from seam images which could demonstrate that the proposed method shows a high detection precision and good robustness against the strong image noise interference from welding process.

关键词： Welding Laser noise Sensors Robots Laser modes Feature extraction Lasers Deep network architecture image denoising seam tracking robot welding structured light vision

来源：评论

学校读者我要写书评

暂无评论

PLE-Net: Automatic power line extraction method using deep learning from aerial images

引用

EXPERT systemS WITH APPLICATIONS 2022年第0期198卷 1页

Automatic power line extraction is a crucial task for the safe navigation of inspection robots. Nevertheless, power lines are always against complicated natural backgrounds which bring a certain challenge for accurate power line extraction. Meanwhile, the power lines always occupy a minimal portion image pixels in the aerial images compared with backgrounds which causes serious class imbalance issue. Therefore, the robust and accurate power line segmentation from aerial images is one of the most frequently stated problems faced with these factors. Recently, the deep learning has got wide applications on different segmentation tasks with effective contextual feature generation ability. However, these methods show poor ability on the samples with class imbalance due to insufficient process of local contextual features. To address these issues, combined with the encoder-decoder framework, a novel power line extraction network (PLE-Net) is proposed in this paper to construct an end-to-end attention-based segmentation method for automatic power line extraction from aerial images with a self-attention block and a multi-scale feature enhance block. To capture rich contextual relationships from local feature maps, a feature enhance block is proposed for multi-scale feature expression. And a self-attention block is proposed to embed into the proposed segmentation network to emphasize the regions about power lines. Further, the hybrid loss function with binary cross-entropy (BCE) and Dice is set as the loss function to address the class imbalance issue. Combined with the public datasets of power lines, the proposed segmentation network shows a better segmentation performance on vision images and infrared images through the ablation analysis and comparison experiments.

关键词： Power line extraction Deep architecture Image segmentation Multi-scale attention module

来源：评论

学校读者我要写书评

暂无评论

A Precise Calibration Method for the Underwater Binocular Camera With Unknown Refractive Index

引用

IEEE SENSORS JOURNAL 2023年第2期23卷 1565-1574页

作者： Yang, Huizhen Tian, Shifei Fan, Junfeng Han, Yanqing Zhou, Chao Ou, Yaming Northwestern Polytech Univ Sch Marine Sci & Technol Xian 710072 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Army Acad Armored Force Dept Basic Course Beijing 100072 Peoples R China

Underwater camera calibration usually needs to solve an accurate refraction model to improve the calibration accuracy. In the refraction model, there are some key parameters: 1) the distance from the optical center to the refraction plane;2) the thickness of the housing;3) the normal vector of the refraction plane;and 4) the refractive index of the medium. In many cases, the refractive index is often assumed to be a known parameter. However, the refractive index of water will change due to environmental changes such as temperature, salinity, or depth. Inaccurate refractive index will seriously affect the measurement accuracy. Therefore, a precise calibration method for the underwater binocular camera with unknown refractive index is proposed. All the unknown parameters of the refraction model, including the refractive index of the medium, are solved by the nondominated sorting genetic algorithm (NSGA)-III. The parameter solving process is regarded as a multiobjective optimization problem, and the optimization objectives are determined by using the relative position and the actual position relationship of the checkerboard corners. Experiments show that the proposed method can greatly improve the 3-D measurement accuracy of binocular camera in unknown waters.

关键词： Refractive camera model underwater calibration underwater measurement underwater stereo vision

来源：评论

学校读者我要写书评

暂无评论

A Novel Ultrasound Robot With Force/Torque Measurement and control for Safe and Efficient Scanning

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2023年 72卷 1页

作者： Bao, Xianqiang Wang, Shuangyi Zheng, Lingling Housden, Richard James Hajnal, Joseph V. Rhode, Kawal Kings Coll London Sch Biomed Engn & Imaging Sci London SE1 7EH England Chinese Acad Sci Inst Automation State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Kagawa Univ Fac Engn & Design Takamatsu 7610396 Japan

Medical ultrasound (US) is of increasing importance in medical diagnosis and intraoperative assistance and possesses great potential advantages when integrated with robotics. However, some concerns, including the operation efficiency, operation safety, image quality, and comfort of patients, remain after introducing robotics into medical ultrasound. In this article, an ultrasound robot integrating a force control mechanism, force/torque measurement mechanism, and online adjustment method is proposed to overcome the current limitations. The ultrasound robot can measure operating forces and torques, provide adjustable constant operating forces, eliminate great operating forces introduced by accidental operations, and achieve various scanning depths based on clinical requirements. The proposed ultrasound robot would potentially facilitate sonographers to find the targets quickly, improve operation safety and efficiency, and decrease patients' discomfort. Simulations and experiments were carried out to evaluate the performance of the ultrasound robot. Experimental results show that the proposed ultrasound robot is able to detect operating force in the z-direction and torques around the x- and y-directions with errors of 3.53%, 6.68%, and 6.11% F.S., maintain the constant operating force with errors of less than 0.57 N, and achieve various scanning depths for target searching and imaging. This proposed ultrasound robot has good performance and would potentially be used in medical ultrasound.

关键词： Ultrasonic imaging Robots Force Force measurement Robot sensing systems Ultrasonic variables measurement Probes operation safety robotic ultrasonography ultrasound image ultrasound robot

来源：评论

学校读者我要写书评

暂无评论

Decoupled Metric Network for Single-Stage Few-Shot Object Detection

引用

IEEE TRANSACTIONS ON CYBERNETICS 2023年第1期53卷 514-525页

作者： Lu, Yue Chen, Xingyu Wu, Zhengxing Yu, Junzhi Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China Kuaishou Technol Ytech Beijing 100085 Peoples R China Peking Univ Coll Engn BIC ESAT Dept Adv Mfg & RobotState Key Lab Turbulence & C Beijing 100871 Peoples R China

Within the last few years, great efforts have been made to study few-shot learning. Although general object detection is advancing at a rapid pace, few-shot detection remains a very challenging problem. In this work, we propose a novel decoupled metric network (DMNet) for single-stage few-shot object detection. We design a decoupled representation transformation (DRT) and an image-level distance metric learning (IDML) to solve the few-shot detection problem. The DRT can eliminate the adverse effect of handcrafted prior knowledge by predicting objectness and anchor shape. Meanwhile, to alleviate the problem of representation disagreement between classification and location (i.e., translational invariance versus translational variance), the DRT adopts a decoupled manner to generate adaptive representations so that the model is easier to learn from only a few training data. As for a few-shot classification in the detection task, we design an IDML tailored to enhance the generalization ability. This module can perform metric learning for the whole visual feature, so it can be more efficient than traditional DML due to the merit of parallel inference for multiobjects. Based on the DRT and IDML, our DMNet efficiently realizes a novel paradigm for few-shot detection, called single-stage metric detection. Experiments are conducted on the PASCAL VOC dataset and the MS COCO dataset. As a result, our method achieves state-of-the-art performance in few-shot object detection. The codes are available at https://***/yrqs/DMNet.

关键词： Object detection Feature extraction Training Head Task analysis Shape Measurement Computer vision deep learning few-shot learning object detection

来源：评论

学校读者我要写书评

暂无评论

Crowd Sensing Intelligence for ITS: Participants, Methods, and Stages

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES

引用

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES 2023年第6期8卷 3541-3546页

作者： Zhao, Yong Hu, Cong Zhu, Zhengqiu Qiu, Sihang Chen, Bin Jiao, Peng Wang, Fei-Yue Natl Univ Def Technol Coll Syst Engn Changsha 410073 Hunan Peoples R China Hunan Inst Adv Technol Changsha 410073 Hunan Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China

The construction of transportation 5.0 or the so-called society-centered intelligent transportation systems (ITS) has aroused higher requirements for the intelligent sensing capability to seamlessly integrate Cyber-Physical-Social systems (CPSS). Crowd Sensing Intelligence (CSI), as a promising paradigm, leverages the collective intelligence of heterogeneous sensing resources to gather data and information from CPSS. Our first Distributed/Decentralized Hybrid Workshop on Crowd Sensing Intelligence (DHW-CSI) has been focused on principles and high-level processes of organizing and operating CSI. This letter reports the discussion results of the second DHW-CSI addressing the participants, methods, and stages of CSI for ITS. We categorized sensing participants into three kinds, i.e., biological, digital, and robotic. Then we summarized three methods to enable sensing intelligence, i.e., foundation models, scenarios engineering, and human-oriented operating systems. Finally, we anticipated that the progression of CSI will experience three stages, from algorithmic intelligence to linguistic intelligence, and eventually to imaginative intelligence.

关键词： Crowd sensing intelligence intelligence transportation systems cyber-physical-social systems

来源：评论

学校读者我要写书评

暂无评论

Time-/Event-Triggered Adaptive Neural Asymptotic Tracking control of Nonlinear Interconnected systems With Unmodeled Dynamics and Prescribed Performance

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING systemS 2023年第9期34卷 6557-6567页

作者： Cheng, Ting-Ting Niu, Ben Zhang, Jia-Ming Wang, Ding Wang, Zhen-Hua Shandong Normal Univ Sch Informat Sci & Engn Jinan 250014 Shandong Peoples R China Beijing Univ Technol Fac Informat Technol Beijing 100124 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China

This article proposes two adaptive asymptotic tracking control schemes for a class of interconnected systems with unmodeled dynamics and prescribed performance. By applying an inherent property of radial basis function (RBF) neural networks (NNs), the design difficulties aroused from the unknown interactions among subsystems and unmodeled dynamics are overcome. Then, in order to ensure that the tracking errors can be suppressed in the specified range, the constrained control problem is transformed into the stabilization problem by using an auxiliary function. Based on the adaptive backstepping method, a time-triggered controller is constructed. It is proven that under the framework of Barbalat's lemma, all the variables in the closed-loop system are bounded and the tracking errors are further ensured to converge to zero asymptotically. Furthermore, the event-triggered strategy with a variable threshold is adopted to make more precise control such that the better system performance can be obtained, which reduces the system communication burden under the condition of limited communication resources. Finally, an illustrative example is provided to demonstrate the effectiveness of the proposed control scheme.

关键词： Interconnected systems Artificial neural networks Closed loop systems Multi-agent systems Asymptotic stability Switches Stability criteria Asymptotic tracking control event-triggered control neural networks (NNs) nonlinear interconnected systems prescribed performance control (PPC) unmodeled dynamics

来源：评论

学校读者我要写书评

暂无评论

SAPM: Self-Adaptive Parallel Manipulator With Pose and Force Adjustment for Robotic Ultrasonography

引用

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS 2023年第10期70卷 10333-10343页

作者： Bao, Xianqiang Wang, Shuangyi Zheng, Lingling Housden, Richard James Hajnal, Joseph Rhode, Kawal Kings Coll London Sch Biomed Engn & Imaging Sci London SE1 7EH England Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Kagawa Univ Fac Engn & Design Takamatsu 7610396 Japan

Robotic ultrasonography potentially acts as an essential aid to medical diagnosis. To overcome the limitations in robotic ultrasonography, in this article, we proposed a novel self-adaptive parallel manipulator (SAPM) that can automatically adjust the ultrasound (US) probe pose to adapt to various contours of scanned areas, provide approximate constant operating forces/torques, achieve mechanical measurement, and cushion undesired produced forces. A novel parallel adjustment mechanism is proposed to attain automatic pose adjustment with 3 degrees of freedom (DOFs). This mechanism enables the US probe to adapt to different scanned areas and to perform the scanning with approximate constant forces and torques. Besides, we present a mechanical measurement and safety protection method that can be integrated into the SAPM and used as operation status monitoring and early warning during scanning procedures by capturing operating forces and torques. Experiments were carried out to calibrate the measurement and buffer units and evaluate the performance of the SAPM. Experimental results show the ability of the SAPM to provide 3 DOFs motion and operating force/torque measurement and automatically adjust the US probe pose to capture US images of equally good quality compared to a manual sonographer scan. Moreover, it has characteristics similar to soft robots that could significantly improve operation safety, and could be extended to some other engineering or medical applications.

关键词： Probes Robots Manipulators Force Safety Force measurement Ultrasonography Automatic pose adjustment constant forces torques mechanical measurement operation safety robotic ultrasonography

来源：评论

学校读者我要写书评

暂无评论

Policy generation network for zero-shot policy learning

引用

COMPUTATIONAL INTELLIGENCE 2023年第5期39卷 707-733页

作者： Qian, Yiming Zhang, Fengyi Liu, Zhiyong Univ Chinese Acad Sci UCAS Sch Artificial Intelligence Beijing 100049 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing Peoples R China Chinese Acad Sci Cloud Comp Ctr Dongguan Guangdong Peoples R China

Lifelong reinforcement learning is able to continually accumulate shared knowledge by estimating the inter-task relationships based on training data for the learned tasks in order to accelerate learning for new tasks by knowledge reuse. The existing methods employ a linear model to represent the inter-task relationships by incorporating task features in order to accomplish a new task without any learning. But these methods may be ineffective for general scenarios, where linear models build inter-task relationships from low-dimensional task features to high-dimensional policy parameters space. Also, the deficiency of calculating errors from objective function may arise in the lifelong reinforcement learning process when some errors of policy parameters restrain others due to inter-parameter correlation. In this paper, we develop a policy generation network that nonlinearly models the inter-task relationships by mapping low-dimensional task features to the high-dimensional policy parameters, in order to represent the shared knowledge more effectively. At the same time, we propose a novel objective function of lifelong reinforcement learning to relieve the deficiency of calculating errors by adding weight constraints for errors. We empirically demonstrate that our method improves the zero-shot policy performance across a variety of dynamical systems.

关键词： knowledge representation lifelong reinforcement learning zero-shot policy generation

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：