检索结果-内蒙古大学图书馆

Image Processing, Computer Vision and Machine Learning (ICICML), International Conference on

作者： Junkai Wang Yu Zhang Ruowu Wu Lianlei Lin State Key Laboratory of Complex Electromagnetic Environment Effects on Electronics and Information System Luoyang China School of Electronics and Information Engineering Harbin Institute of Technology Harbin China

ISBN: (数字)9798350355413

ISBN: (纸本)9798350355420

Various natural disasters are closely related to the El Niño - Southern Oscillation (ENSO), the stable generation of accurate sea surface temperature anomaly (SSTA) data is key to exploring the occurrence patterns of ENSO and preventing disasters. In this study, we introduces the Separated Attention Data Generation (SADG) model, which efficiently generates SSTA data by capturing the long-range spatiotemporal relationship. SADG achieves high performance while reducing computational complexity compared to standard self-attention based models, generates up to 18 months of accurate SSTA data. Experiments on CMIP datasets show that its performance is better than other models on various lead times.

关键词： Accuracy Computational modeling Disasters Data collection Predictive models Data models Spatiotemporal phenomena Sea surface temperature Standards Oscillators

来源：评论

学校读者我要写书评

暂无评论

China’s larch stock volume estimation using Sentinel-2 and LiDAR data

引用

Geo-Spatial Information Science 2023年第3期26卷 392-405页

作者： Tao Yu Yong Pang Xiaojun Liang Wen Jia Yu Bai Yilin Fan Dongsheng Chen Xianzhao Liu Guang Deng Chonggui Li Xiangnan Sun Zhidong Zhang Weiwei Jia Zhonghua Zhao Xiao Wang Institute of Forest Resource Information Techniques Chinese Academy of ForestryBeijingChina Key Laboratory of Forestry Remote Sensing and Information System National Forestry and Grassland AdministrationBeijingChina Key Laboratory of Tree Breeding and Cultivation of State Forestry Administration Research Institute of ForestryChinese Academy of ForestryBeijingChina College of Geomatics Xi’an University of Science and TechnologyXi’anChina Academy of Inventory and Planning National Forestry and Grassland AdministrationBeijingChina College of Forestry Hebei Agricultural UniversityBaodingChina College of Forestry Northeast Forestry UniversityHarbinChina Ecology and Nature Conservation Institute Chinese Academy of ForestryKey Laboratory of Forest Ecology and Environment of National Forestry and Grassland AdministrationBeijingChina

Forest Stock Volume(FSV)is one of the key indicators in forestry resource investigation and management on local,regional,and national *** by the saturation problems of optical satellite remote-sensing imagery in the retrieving of stock volume,and the high cost of Light Detection And Ranging(LiDAR)data,it is still challenging to estimate FSV in a large area using single-sensor remote-sensing *** this paper,a method integrated multispectral satellite imagery and LiDAR data was developed to map stock volume in a large area.A random forest model was adopted to estimate the stock volume of larch forest in China based on the training samples from the Airborne Laser Scanning(ALS)-derived stock volume and corresponding Sentinel-2 *** using National Forest Inventory(NFI)data,ALS-derived stock volume and ground investigation data demonstrated that the estimated stock volume had a high accuracy(R2=0.59,RMSE=59.69 m^(3)/ha,MD=39.96 m^(3)/ha when validated with NFI data;R2 ranged from 0.77 to 0.85,RMSE ranged from 38.68 m^(3)/ha to 67.38 m^(3)/ha,MD ranged from 24.90 m^(3)/ha to 37.27 m^(3)/ha when validated with ALS stock volume;R2=0.42,RMSE=79.10 m^(3)/ha,MD=62.06 m^(3)/ha when validated with field investigation data).Results of this paper indicated the applicability of estimating stock volume of larch forest in a large area by combining Sentinel-2 data and airborne LiDAR data.

关键词： Sentinel-2 LiDAR stock volume larch validation

来源：评论

学校读者我要写书评

暂无评论

Sim-to-Real Grasp Detection with Global-to-Local RGB-D Adaptation

Sim-to-Real Grasp Detection with Global-to-Local RGB-D Adapt...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Haoxiang Ma Ran Qin Modi Shi Boyang Gao Di Huang State Key Laboratory of Software Development Environment School of Computer Science and Engineering Beihang University Beijing China Geometry Robotics and the School of Computer Science and Technology Harbin Institute of Technology Harbin China

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

This paper focuses on the sim-to-real issue of RGB-D grasp detection and formulates it as a domain adaptation problem. In this case, we present a global-to-local method to address hybrid domain gaps in RGB and depth data and insufficient multi-modal feature alignment. First, a self-supervised rotation pre-training strategy is adopted to deliver robust initialization for RGB and depth networks. We then propose a global-to-local alignment pipeline with individual global domain classifiers for scene features of RGB and depth images as well as a local one specifically working for grasp features in the two modalities. In particular, we propose a grasp prototype adaptation module, which aims to facilitate fine-grained local feature alignment by dynamically updating and matching the grasp prototypes from the simulation and real-world scenarios throughout the training process. Due to such designs, the proposed method substantially reduces the domain shift and thus leads to consistent performance improvements. Extensive experiments are conducted on the GraspNet-Planar benchmark and physical environment, and superior results are achieved which demonstrate the effectiveness of our method. Code is available at https://***/mahaoxiang822/GL-MSDA.

关键词： Training Codes Pipelines Prototypes Benchmark testing Robustness robotics and automation

来源：评论

学校读者我要写书评

暂无评论

Sim-to-Real Grasp Detection with Global-to-Local RGB-D Adaptation

arXiv

引用

arXiv 2024年

作者： Ma, Haoxiang Qin, Ran Shi, Modi Gao, Boyang Huang, Di The State Key Laboratory of Software Development Environment School of Computer Science and Engineering Beihang University Beijing China The Geometry Robotics and the School of Computer Science and Technology Harbin Institute of Technology Harbin China

关键词：

来源：评论

学校读者我要写书评

暂无评论

I2EKF-LO: A Dual-Iteration Extended Kalman Filter Based LiDAR Odometry

I2EKF-LO: A Dual-Iteration Extended Kalman Filter Based LiDA...

引用

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

作者： Wenlu Yu Jie Xu Chengwei Zhao Lijun Zhao Thien-Minh Nguyen Shenghai Yuan Mingming Bai Lihua Xie State Key Laboratory of Robotics and Systems Harbin Institute of Technology Harbin China School of Electrical and Electronic Engineering Nanyang Technological University Singapore Hangzhou Guochen Robot Technology Company Limited Hangzhou China State Key Laboratory of Industrial Control Technology Zhejiang University Hangzhou China

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

LiDAR odometry is a pivotal technology in the fields of autonomous driving and autonomous mobile robotics. However, most of the current works focus on nonlinear optimization methods, and still existing many challenges in using the traditional Iterative Extended Kalman Filter (IEKF) framework to tackle the problem: IEKF only iterates over the observation equation, relying on a rough estimate of the initial state, which is insufficient to fully eliminate motion distortion in the input point cloud; the system process noise is difficult to be determined during state estimation of the complex motions; and the varying motion models across different sensor carriers. To address these issues, we propose the Dual-Iteration Extended Kalman Filter (I 2 EKF) and the LiDAR odometry based on I 2 EKF (I 2 EKF-LO). This approach not only iterates over the observation equation but also leverages state updates to iteratively mitigate motion distortion in LiDAR point clouds. Moreover, it dynamically adjusts process noise based on the confidence level of prior predictions during state estimation and establishes motion models for different sensor carriers to achieve accurate and efficient state estimation. Comprehensive experiments demonstrate that I 2 EKF-LO achieves outstanding levels of accuracy and computational efficiency in the realm of LiDAR odometry. Additionally, to foster community development, our code is open-sourced. 1

关键词： Point cloud compression Laser radar Accuracy Nonlinear distortion Noise Robot sensing systems Mathematical models Odometry Kalman filters state estimation

来源：评论

学校读者我要写书评

暂无评论

Moobot: a miniature origami omnidirectional jumping robot with high trajectory accuracy

TechRxiv

引用

TechRxiv 2022年

作者： Tang, Lingqi Li, Yao Li, Bing School of Mechanical Engineering and Automation Harbin Institute of Technology Shenzhen Shenzhen518055 China State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin150001 China

The jumping trajectory control is important for the jumping robots to overcome different obstacles. However, the trajectory control of a micro jumping robot is not easy. No insect-scale jumping robots have demonstrated the precise trajectory following yet. We proposed a miniature origami omnidirectional jumping robot (Moobot) that achieved the feedback control of jumping trajectory. The jumping of the robot is based on a four-bar origami structure and the SMA actuators. Unlike the tethered micro robots that powered externally, we integrated a miniature control board with LiPo battery on the robot for remote control. The overall system measured 6 g in mass, 5.5 cm in length and 2cm in height. Two strain gauges and an IMU sensor were used on the robot for the feedback control of the jumping force and take-off angle, respectively. Meanwhile, a reliable jumping mechanism and a novel supporting leg were designed to improve the jumping stability and trajectory accuracy. The standard deviation of the jumping distances was less than 1.4 cm, which is greatly improved from all previous micro jumping robots. Furthermore, we integrated the yaw control and conducted the omnidirectional jumping experiments. The results revealed that Moobot could climb different platforms from all directions precisely. © 2022, CC BY.

关键词： Feedback control

来源：评论

学校读者我要写书评

暂无评论

Exploring Supervised Contrastive Learning for Skeleton-Based Temporal Action Segmentation

引用

IEEE Transactions on Cognitive and Developmental systems 2025年

作者： Chen, Bowen Ji, Haoyu Ma, Hanwei Lin, Ruihan Nie, Wei Ren, Weihong Wang, Zhiyong Liu, Honghai Harbin Institute of Technology Shenzhen State Key Laboratory of Robotics and Systems Shenzhen518055 China University of Portsmouth School of Computing PortsmouthPO1 3HE United Kingdom

Skeleton-based temporal action segmentation (STAS) takes long human skeleton sequences as input and predicts action categories at the frame level. Current STAS methods primarily use frame-wise cross-entropy loss, which focuses only on the relationships among one-hot label logits and overlooks the importance of the quality of frame-wise representations. To this end, we propose a novel framework called Supervised Contrastive Skeleton-Based Temporal Action Segmentation (SCSAS) that optimizes representation learning. Specifically, our framework constructs a frame-level embedding space using a simple projection head and optimizes this space using three novel contrastive losses. These losses enhance the semantic relationships at the frame and segment levels by pulling together representations of the same activities and pushing apart those of different actions. Moreover, we introduce a confidence-based hard anchor sampling strategy to enhance the efficiency of contrastive learning. Finally, a boundary refinement module is also introduced to fully exploit the advantages of optimized representation. Our method seamlessly integrates with existing STAS methods and consistently enhances their performance across various datasets, without additional inference costs. With the incorporation of the boundary refinement branch, early STAS methods even achieve performance comparable to the state-of-the-art methods. © 2016 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Combined studies of surface evolution and crack healing for the suppression of negative factors during CO_(2) laser repairing of fused silica

引用

Chinese Optics Letters 2021年第4期19卷 44-49页

作者： Chao Tan Linjie Zhao Mingjun Chen Jian Cheng Zhaoyang Yin Qi Liu Hao Yang Wei Liao State Key Laboratory of Robotics and System Harbin Institute of TechnologyHarbin 150001China Research Center of Laser Fusion China Academy of Engineering PhysicsMianyang 621900China

In order to reveal the evolution mechanism of repaired morphology and the material's migration mechanism on the crack surface in the process of CO_(2) laser repairing surface damage of fused silica optics, two multi-physics coupling mathematical models with different scales are developed, respectively. The physical problems, such as heat and mass transfer,material phase transition, melt flow, evaporation removal, and crack healing, are analyzed. Studies show that material ablation and the gasification recoil pressure accompanying the material splash are the leading factors in forming the Gaussian crater with a raised rim feature. The use of low-power lasers for a long time can fully melt the material around the crack before healing, which can greatly reduce the size of the residual air layer. Combined with the experimental research, the methods to suppress the negative factors(e.g., raised rim, deposited debris, air bubbles) in the CO_(2) laser repairing process are proposed.

关键词： CO_(2)laser repairing fused silica crack healing surface evolution

来源：评论

学校读者我要写书评

暂无评论

Unidirectional acoustic parametric amplification in space-time modulated membrane system

Unidirectional acoustic parametric amplification in space-ti...

引用

ASME 2020 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC-CIE 2020

作者： Zhu, Xiaohui Li, Longqiu Zhang, Guangyu State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin Heilongjiang China

ISBN: (纸本)9780791883907

Space-time modulation puts a new twist on controlling wave behaviors and opens the door for unprecedented wave manipulation capabilities. Here we propose a space-time modulated membrane system aiming to realize the unidirectional parametric amplification. Two different approaches, continuous model and transfer matrix method, are applied to analyze the acoustic response of the system under a monochromatic incidence. In the proposed space-time modulated membrane system, these two methods are in good agreement with each other. Results show that under a monochromatic incidence from one end of the system, a harmonic is generated, and both the incident wave and generated wave exponentially grow along with the system. Our work demonstrates a metamaterial realization of unidirectional acoustic parametric amplifier via space-time modulated membranes, which offers a design platform for a number of applications in sensing, imaging and communication. Copyright © 2020 ASME.

关键词： Transfer matrix method

来源：评论

学校读者我要写书评

暂无评论

A Biomimetic impedance controller for Robotic Hand Variable Stiffness Grasping 17

A Biomimetic impedance controller for Robotic Hand Variable ...

引用

17th IEEE International Conference on Mechatronics and Automation, ICMA 2020

作者： Liu, Bingchen Jiang, Li Fan, Shaowei Li, Chongyang Harbin Institute of Technology State Key Laboratory of Robotics and System Harbin150080 China

ISBN: (纸本)9781728164151

The accuracy of human hand grasp is due to many reasons, e.g. the feedbacks of force, temperature, pressure, and position, etc. It is impossible to mimic this complicated system for a robotic hand. Therefore, simplification of the robotic control system with relative accurate operations is necessary. In this study, a biomimetic impedance controller was utilized on a robotic hand, where the control parameters were based on a biomechanical model of the human hand. The relationship between the grasp force and surface electromyography (sEMG) was measured, and the impedance of the joint was calculated through a series of dynamic vibration tests subsequently. The effectiveness of the biomimetic impedance controller is verified through human operating the robotic hand reproducing a variety of grasping tasks. © 2020 IEEE.

关键词： Biomimetics

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：