检索结果-内蒙古大学图书馆

SSRN 2024年

作者： Zheng, Chaobing Yao, Yao Ying, Wenjian Wu, Shiqian Institute of Robotics and Intelligent Systems School of Information Science and Engineering Wuhan University of Science and Technology 947 Heping Avenue Wuhan430081 China Division of Mechanical Manufacturing and Intelligent Transportation Beijing Institute of Metrology Beijing100029 China College of Weapon Naval University of Engineering Wuhan430033 China

It is challenging to remove rain-steaks from a single image because the rain steaks are spatially varying in the rainy image. While data-driven rain removal algorithms have demonstrated promising results, they still face limitations, such as a heavy reliance on data and limited interpretability. This paper introduces a novel approach for single-image de-raining guided by Fourier Transform prior knowledge. Our method leverages inherent frequency domain information to effectively reduce rain streaks and restore image clarity. Initially, using Fourier transform to decompose the rainy image into amplitude and phase components, where raindrops are predominantly present in the amplitude image. Subsequently, data-driven algorithms are applied separately to process the amplitude and phase components. The Fourier inverse transform is then utilised to obtain enhanced features with improved clarity. In the final step, a multi-scale neural network incorporating a different scale attention mechanism is designed to enhance the processed features, thereby improving the algorithm's robustness. Experimental results demonstrate that the proposed algorithm significantly outperforms state-of-the-art methods, both qualitatively and quantitatively. This innovative approach combines the power of the Fourier Transform with data-driven techniques, providing a more interpretable and effective solution for single-image de-raining. © 2024, The Authors. All rights reserved.

关键词： Rain

来源：评论

学校读者我要写书评

暂无评论

Flapping-Wing Robot Achieves Multi-Domain Locomotion Using Transformable Wheel Mechanism

Flapping-Wing Robot Achieves Multi-Domain Locomotion Using T...

引用

IEEE International Conference on robotics and Biomimetics

作者： Xiangxiao Chai Shihua Wang Erzhen Pan Wenfu Xu School of Mechanical Engineering and Automation Harbin Institute of Technology (Shenzhen) Shenzhen China Key University Laboratory of Mechanism & Machine Theory and Intelligent Unmanned Systems of Guangdong Shenzhen China State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

ISBN: (数字)9798331509644

ISBN: (纸本)9798331509651

The difficulty of moving from aquatic to aerial environment is one of the major challenges limiting the development of trans-medium multi-domain locomotion vehicles. In this paper, the concept of the transformable wheel, a feature commonly found in autonomous land vehicles (AL V s), is introduced into the design. Additionally, a flapping-wing robot equipped with multi-domain locomotion capability, which can not only realize aerial flight and surface navigation on water, but also ascend from water to land for subsequent takeoff is presented. The paper first introduces the system design of the robot and proposes the corresponding multi-domain locomotion strategy. Subsequently, the design of the transformable wheel mechanism is described in detail and the force analysis for the water-exit is conducted. The feasibility of the proposed strategy is experimentally validated, confirming the prototype's multi-domain locomotion performance finally. This research explores the feasibility of introducing the transformable wheel mechanism into trans-medium multi-domain vehicle platforms and provides a design example.

关键词： Legged locomotion Torque Limiting Navigation Propellers Roads Wheels Prototypes Motors System analysis and design

来源：评论

学校读者我要写书评

暂无评论

3D Facial Expressions through Analysis-by-Neural-Synthesis

3D Facial Expressions through Analysis-by-Neural-Synthesis

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： George Retsinas Panagiotis P. Filntisis Radek Daněček Victoria F. Abrevaya Anastasios Roussos Timo Bolkarr Petros Maragos Institute of Robotics Athena Research Center Maroussi Greece MPI for Intelligent Systems Tübingen Germany Institute of Computer Science (ICS) Foundation for Research & Technology - Hellas (FORTH) Greece School of Electrical & Computer Engineering National Technical University of Athens Greece

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

While existing methods for 3D face reconstruction from in-the-wild images excel at recovering the overall face shape, they commonly miss subtle, extreme, asymmetric, or rarely observed expressions. We improve upon these meth-ods with SMIRK (Spatial Modeling for Image-based Reconstruction of Kinesics), which faithfully reconstructs expres-sive 3D faces from images. We identify two key limitations in existing methods: shortcomings in their self-supervised training formulation, and a lack of expression diversity in the training images. For training, most methods employ differentiable rendering to compare a predicted face mesh with the input image, along with a plethora of additional loss functions. This differentiable rendering loss not only has to provide supervision to optimize for 3D face geom-etry, camera, albedo, and lighting, which is an ill-posed optimization problem, but the domain gap between ren-dering and input image further hinders the learning pro-cess. Instead, SMIRK replaces the differentiable rendering with a neural rendering module that, given the ren-dered predicted mesh geometry, and sparsely sampled pix-els of the input image, generates a face image. As the neural rendering gets color information from sampled im-age pixels, supervising with neural rendering-based reconstruction loss can focus solely on the geometry. Further it enables us to generate images of the input identity with varying expressions while training. These are then utilized as input to the reconstruction model and used as supervision with ground truth geometry. This effectively augments the training data and enhances the generalization for di-verse expressions. Our qualitative, quantitative and partic-ularly our perceptual evaluations demonstrate that SMIRK achieves the new state-of-the art performance on accurate expression reconstruction. For our method's source code, demo video and more, please visit our project webpage: https://***/smirk/.

关键词： Training Geometry Solid modeling Three-dimensional displays Accuracy Face recognition Training data

来源：评论

学校读者我要写书评

暂无评论

SLAM-Based Joint Calibration of Differential RSS Sensor Array and Source Localization 49

SLAM-Based Joint Calibration of Differential RSS Sensor Arra...

引用

49th Annual Conference of the IEEE Industrial Electronics Society, IECON 2023

作者： Fu, Linya Qiao, Xu Huang, Shoudong Mao, Guoqiang Lin, Zhiyun Li, Youfu Kong, He Robotics Institute Uni. of Technology Sydney Sydney Australia Xidian University State Key Laboratory of Integrated Services Networks China City University of Hong Kong Department of Mechanical Engineering China Southern Uni. of Sci. and Tech. Shenzhen Key Laboratory of Control Theory and Intelligent Systems Shenzhen China Southern Uni. of Sci. and Tech. Guangdong Prov. Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities Shenzhen China

ISBN: (纸本)9798350331820

Sensor arrays generating differential received signal strength (DRSS) measurements have found many applications in robotics. However, accurate calibration of these sensor arrays remains a challenge. Most existing methods are impractical in that they assume to know signal source positions or certain parameters (i.e., path loss exponent), and try to estimate the others. In this paper, we adopt graph simultaneous localization and mapping (SLAM) as a general framework for jointly estimating the source positions and parameters of the DRSS sensor array. Our contributions are twofold. On the one hand, by using a Fisher information matrix approach, we conduct a systematic observability analysis of the corresponding SLAM setup for the calibration problem. On the other hand, we propose an effective procedure to select the initial value which is fed to Levenberg-Marquardt iterations for further improving optimization accuracy and convergence. Extensive simulation and hardware experiments show that the proposed method renders high-quality calibration results. All the codes and data are publicly available at https://***/SUSTech2022/DRSS-sensor-array-calibration. © 2023 IEEE.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

A multiplexed FBG based sensor platform for flow and temperature measurements in the Baltic Sea

A multiplexed FBG based sensor platform for flow and tempera...

引用

2023 European Workshop on Optical Fibre Sensors, EWOFS 2023

作者： Dzipalski, A. Morton, J.A.S. Papachristou, N. Maier, R.R.J. MacPherson, W.N. Ristolainen, A. Kruusmaa, M. Reilent, E. Wolf, B.J. Pirih, P. van Netten, S.M. Shuhova, I. Lips, U. McFarlane, N. Macleod, R. Hendry, M. Sheehy, J. Almoghayer, M. Rojas, N. Davies, G. Hakim, A. Ng, A. Heriot-Watt University Institute of Photonics and Quantum Sciences Edinburgh United Kingdom Tallinn Institute of Technology Centre for Bio Robotics Tallinn Estonia University of Groningen Computer Science and Artificial Intelligence Groningen Netherlands Tallinn Institute of technology Department of Marine Systems Tallinn Estonia Hydrobond Engineering Aberdeen Energy Park Aberdeen United Kingdom Aquatera Ltd The Charles Coulson Building Stromness United Kingdom Orcades Marine Warness Park Kirkwall United Kingdom

ISBN: (纸本)9781510665002

In this paper a multiplexed multi-parameter marinized sensory array is described. This was deployed on the continental slope off the Keri Island marine observatory in the Gulf of Finland (Estonia). The sensor array is made up of 4 measurement stations which are connected in series. Across these measurement stations, a total of 16 temperature sensors, 4 attitude sensors (each consisting of 3 individual fiber sensors) and 16 flow sensors were successfully deployed. They were addressed over 1.1 km via 20 single-mode (SMF-28e) optical fibers contained in a single marine compliant ruggedized umbilical. The bio-inspired fiber optic flow sensors are designed to mimic the behavior of the superficial neuromasts naturally found as part of the lateral line sensory organ present in fish. The sensor is composed of optical fibers inscribed with Fiber Bragg Gratings glued together in a polymer matrix which are then encapsulated in a polyurethane shell. The sensors response has been tested in DC flows in a tow tank and have demonstrated the ability of measuring flow speed from 0.05 ms-1 to 2.5 ms-1. The main aim of the deployment was to demonstrate the capabilities of fiber sensor technology for long-term oceanographic applications. The measurement period described lasted over two months and the sensor system performed well in comparison with data was gathered from commercial instrumentation available. © 2023 SPIE.

关键词： Fiber Bragg gratings

来源：评论

学校读者我要写书评

暂无评论

Dynamic Setpoint Optimization Using Metaheuristic Algorithms for Wastewater Treatment Plants 48

Dynamic Setpoint Optimization Using Metaheuristic Algorithms...

引用

48th Annual Conference of the IEEE Industrial Electronics Society, IECON 2022

作者： Salles, Rodrigo Mendes, Jerome Henggeler Antunes, Carlos Moura, Pedro Dias, Joana University of Coimbra Institute of Systems and Robotics Department of Electrical and Computer Engineering Pólo II Coimbra3030-290 Portugal University of Coimbra Inesc Coimbra Department of Electrical and Computer Engineering Pólo II Coimbra3030-290 Portugal Ctga Park no 70 Taveiro Coimbra3045-487 Portugal

ISBN: (数字)9781665480253

ISBN: (纸本)9781665480253

Wastewater Treatment Plants (WWTPs) are complex structures, with nonlinear characteristics, that have strict quality criteria, and carry out the treatment using various resources, with special emphasis on electricity. The optimization of WWTPs is a necessity to achieve sustainability. This work proposes the use of metaheuristic algorithms to optimize the functioning of a WWTP. Genetic Algorithm, Particle Swarm Optimization, and Simulated Annealing are used to minimize the aeration energy consumption prediction by an artificial neural network. The optimization framework aims dynamically adjusts the reference value for the proportional-integral (PI) controller. The proposed approach was able to dynamically adjust the oxygen pumping requirement and thus reduce the electricity consumption compared to the default PI controller adopted by the Benchmark Simulation Model no. 2 (BSM2). © 2022 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

High-Precision Localization of Mobile Robot Based on Particle Filters Combined with Triangle Matching 7th

High-Precision Localization of Mobile Robot Based on Parti...

引用

7th International Conference on Cognitive systems and Information Processing, ICCSIP 2022

作者： Zhou, Huaidong Tuo, Wanchen Chou, Wusheng Robotics Institute School of Mechanical Engineering and Automation Beihang University Beijing100191 China State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing100191 China

ISBN: (纸本)9789819906161

Robot location is a fundamental technology in the field of autonomous robotics and has been widely studied. However, high-precision localization of the mobile robot is still a serious challenge. To investigate this point further, this paper proposes a method to improve the autonomous robot localization accuracy, which adjusts the number of particles in the AMCL algorithm by fusing the information of LRL. LRL data is used to improve the localization accuracy of mobile platforms and prevent kidnapped robot problems. The algorithm is conducted and verified in a real robot system. The experiment results show that the proposed method can achieve high-precision locations in real-world scenarios, and verify the effectiveness of the improved localization system. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Mobile robots

来源：评论

学校读者我要写书评

暂无评论

Saliency-SLAM: saliency prediction for visual SLAM

Saliency-SLAM: saliency prediction for visual SLAM

引用

Tianjin University-IET Electrical & Information engineering Doctoral Forum (TJU IET 2023)

作者： S. Jin Q. Meng Institute of Robotics and Autonomous Systems Tianjin Key Laboratory of Intelligent Unmanned Swarm Technology and System School of Electrical and Information Engineering Tianjin University Tianjin People's Republic of China

This paper proposes a novel approach for visual simultaneous localization and mapping (vSLAM), called Saliency-SLAM. This method effectively emulates the visual attention mechanism by integrating a saliency prediction model. In contrast to conventional vSLAM methods, which treat all feature points as equally significant, salient feature points are given greater importance in Saliency-SLAM. Therefore, we construct a novel saliency dataset specifically for vSLAM task to train a saliency prediction model by integrating multi-level visual information. Then, we propose to use the salient value of predicted result as the weight of significance for the feature points during the optimization process. Exhaustive experiments have been conducted on the KITTI dataset, demonstrating that our proposed method outperforms existing approaches in terms of localization accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-sensory Olfactory Quadruped Robot for Odor Source Localization*

Multi-sensory Olfactory Quadruped Robot for Odor Source Loca...

引用

IEEE International Conference on Cyborg and Bionic systems (CBS)

作者： Yu He Lei Cheng YaDuo Pan YuAo Li DuanChu Wang Han Zheng School of Information Science and Engineering/School of Artificial Intelligence Wuhan University of Science and Technology Wuhan China Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology Wuhan China

With the rapid development of industrialization, hazardous chemical gas leakage accidents not only cause huge losses to people and economy, but also cause great damage to the environment. Research into how to efficiently, safely and accurately search for and locate sources of toxic, hazardous, explosive, flammable and other dangerous chemical gas leaks has become an urgent and realistic need. In this work, we focused on promoting the application scenarios of mobile robot-based odor source localization technology and proposed an olfactory quadruped robot capable of odor source localization tasks in complex environments. Compared to traditional wheeled robots, olfactory quadruped robot are more adaptive and eco-friendly. The robot is equipped with a variety of sensors and has a rich perception of complex environments, enabling it to complete complex tasks such as mapping, navigation and odor source localization in unstructured terrain.

关键词： Location awareness Navigation Olfactory Robot sensing systems Reliability engineering Sensors Mobile robots

来源：评论

学校读者我要写书评

暂无评论

Ground-optimized SLAM with Hierarchical Loop Closure Detection in Large-scale Environment

Ground-optimized SLAM with Hierarchical Loop Closure Detecti...

引用

International Conference on Intelligent Transportation

作者： Huilin Yin Mina Sun Linchuan Zhang Jun Yan Johannes Betz College of Electronics and Information Engineering Tongji University Shanghai China Professorship of Autonomous Vehicle Systems TUM School of Engineering and Design Technical University Munich Garching Germany Munich Institute of Robotics and Machine Intelligence (MIRMI)

Accurate and reliable localization and mapping are crucial prerequisites for autonomous driving to achieve path planning. However, in large-scale complex dynamic environments, the traditional LiDAR loop closure detection methods that rely on radius search can easily result in false negatives, leading to the inability to correct accumulated errors effectively. To address this issue, a hierarchical LiDAR descriptor loop closure detection strategy is proposed in this paper, which detects invalid loop closures and has good viewpoint invariance. We integrate this strategy into an advanced LiDAR Inertial tightly coupled SLAM framework. In addition, to reduce the drift in the vertical direction during mapping, we introduce a ground marking algorithm and construct corresponding ground constraints in the back-end optimization. Our proposed method is evaluated on the MulRan dataset, and the experimental results show that our method could achieve lower accumulated errors than competing methods.

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：