检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Shi, Hao Pang, Chengshan Zhang, Jiaming Yang, Kailun Wu, Yuhao Ni, Huajian Lin, Yining Stiefelhagen, Rainer Wang, Kaiwei The State Key Laboratory of Extreme Photonics and Instrumentation Zhejiang University Hangzhou310027 China The Institute for Anthropomatics and Robotics Karlsruhe Institute of Technology Karlsruhe76131 Germany The Department of Engineering Science University of Oxford OxfordOX1 3PJ United Kingdom The School of Robotics Hunan University Changsha410012 China The National Engineering Research Center of Robot Visual Perception and Control Technology Hunan University Changsha410082 China Shanghai SUPREMIND Technology Co. Ltd. Shanghai201210 China

Roadside camera-driven 3D object detection is a crucial task in intelligent transportation systems, which extends the perception range beyond the limitations of vision-centric vehicles and enhances road safety. While previous studies have limitations in using only depth or height information, we find both depth and height matter and they are in fact complementary. The depth feature encompasses precise geometric cues, whereas the height feature is primarily focused on distinguishing between various categories of height intervals, essentially providing semantic context. This insight motivates the development of Complementary-BEV (CoBEV), a novel end-to-end monocular 3D object detection framework that integrates depth and height to construct robust BEV representations. In essence, CoBEV estimates each pixel's depth and height distribution and lifts the camera features into 3D space for lateral fusion using the newly proposed two-stage complementary feature selection (CFS) module. A BEV feature distillation framework is also seamlessly integrated to further enhance the detection accuracy from the prior knowledge of the fusion-modal CoBEV teacher. We conduct extensive experiments on the public 3D detection benchmarks of roadside camera-based DAIR-V2X-I and Rope3D, as well as the private Supremind-Road dataset, demonstrating that CoBEV not only achieves the accuracy of the new state-of-the-art, but also significantly advances the robustness of previous methods in challenging long-distance scenarios and noisy camera disturbance, and enhances generalization by a large margin in heterologous settings with drastic changes in scene and camera parameters. For the first time, the vehicle AP score of a camera model reaches 80% on DAIR-V2X-I in terms of easy mode. The source code will be made publicly available at CoBEV. Copyright © 2023, The Authors. All rights reserved.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

TransKD: Transformer Knowledge Distillation for Efficient Semantic Segmentation

arXiv

引用

arXiv 2022年

作者： Liu, Ruiping Yang, Kailun Roitberg, Alina Zhang, Jiaming Peng, Kunyu Liu, Huayao Wang, Yaonan Stiefelhagen, Rainer The Institute for Anthropomatics and Robotics Karlsruhe Institute of Technology Karlsruhe76131 Germany The School of Robotics The National Engineering Laboratory of Robot Visual Perception and Control Technology Hunan University Changsha410082 China The Institute for Artificial Intelligence The University of Stuttgart Stuttgart70569 Germany The Institute for Visual Computing ETH Zurich Zurich8092 Switzerland NIO Shanghai201804 China

Semantic segmentation benchmarks in the realm of autonomous driving are dominated by large pre-trained transformers, yet their widespread adoption is impeded by substantial computational costs and prolonged training durations. To lift this constraint, we look at efficient semantic segmentation from a perspective of comprehensive knowledge distillation and aim to bridge the gap between multi-source knowledge extractions and transformer-specific patch embeddings. We put forward the Transformer-based Knowledge Distillation (TransKD) framework which learns compact student transformers by distilling both feature maps and patch embeddings of large teacher transformers, bypassing the long pre-training process and reducing the FLOPs by >85.0%. Specifically, we propose two fundamental modules to realize feature map distillation and patch embedding distillation, respectively: (1) Cross Selective Fusion (CSF) enables knowledge transfer between cross-stage features via channel attention and feature map distillation within hierarchical transformers;(2) Patch Embedding Alignment (PEA) performs dimensional transformation within the patchifying process to facilitate the patch embedding distillation. Furthermore, we introduce two optimization modules to enhance the patch embedding distillation from different perspectives: (1) Global-Local Context Mixer (GL-Mixer) extracts both global and local information of a representative embedding;(2) Embedding Assistant (EA) acts as an embedding method to seamlessly bridge teacher and student models with the teacher’s number of channels. Experiments on Cityscapes, ACDC, NYUv2, and Pascal VOC2012 datasets show that TransKD outperforms state-of-the-art distillation frameworks and rivals the time-consuming pre-training method. The source code is publicly available at https://***/RuipingL/TransKD. Copyright © 2022, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Learning compliant grasping and manipulation by teleoperation with adaptive force control

arXiv

引用

arXiv 2021年

作者： Zeng, Chao Li, Shuang Jiang, Yiming Li, Qiang Chen, Zhaopeng Yang, Chenguang Zhang, Jianwei Department of Informatics Universiẗat Hamburg Germany National Engineering Laboratory for Robot Visual Perception and Control Hunan University China Center for Cognitive Interaction Technology Bielefeld University Germany Bristol Robotics Laboratory University of the West of England United Kingdom

In this work, we focus on improving the robot's dexterous capability by exploiting visual sensing and adaptive force control. TeachNet, a vision-based teleoperation learning framework, is exploited to map human hand postures to a multifingered robot hand. We augment TeachNet, which is originally based on an imprecise kinematic mapping and position-only servoing, with a biomimetic learning-based compliance control algorithm for dexterous manipulation tasks. This compliance controller takes the mapped robotic joint angles from TeachNet as the desired goal, computes the desired joint torques. It is derived from a computational model of the biomimetic control strategy in human motor learning, which allows adapting the control variables (impedance and feedforward force) online during the execution of the reference joint angle trajectories. The simultaneous adaptation of the impedance and feedforward profiles enables the robot to interact with the environment in a compliant manner. Our approach has been verified in multiple tasks in physics simulation, i.e., grasping, opening-a-door, turning-a-cap, and touching-a-mouse, and has shown more reliable performances than the existing position control and the fixed-gain-based force control approaches. Copyright © 2021, The Authors. All rights reserved.

关键词： Remote control

来源：评论

学校读者我要写书评

暂无评论

Cleaning robot Operation Decision Based on Causal Reasoning and Attribute Learning*

Cleaning Robot Operation Decision Based on Causal Reasoning ...

引用

2020 IEEE/RSJ International Conference on Intelligent robots and Systems (IROS)

作者： Yapeng Li Dongbo Zhang Feng Yin Ying Zhang Institute of Automation and Electronic Information Xiangtan University Xiangtan China Institute of National Engineering Laboratory of Robot Vision Perception and Control Technology Xiangtan University Xiangtan China

ISBN: (数字)9781728162126

ISBN: (纸本)9781728162133

In order to improve the operation ability of cleaning robots, this paper proposes a decision method for cleaning robot's operation mode. Firstly, we use the hierarchical expression ability of deep network to obtain the attributes of garbage such as state, shape, distribution, size and so on. Then the causal relationship between the attributes and the operation modes can be built by using joint learning of association attributes with depth network model and causal inference. Based on this, a fuzzy inference decision network is designed. With the help of causal analysis, the structure of the decision model is greatly simplified. Compared with conventional fuzzy neural networks, the total parameters of the model are reduced by 2 / 3. The method proposed in this paper imitates the way that human dispose of different types of garbage and has good interpretability. The experimental results verify the effectiveness of the proposed method.

关键词： Training Shape Semantics Fuzzy neural networks Cleaning Cognition robots

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Object Pose Estimation: A Comprehensive Survey

arXiv

引用

arXiv 2024年

作者： Liu, Jian Sun, Wei Yang, Hui Zeng, Zhiwen Liu, Chongpei Zheng, Jin Liu, Xingyu Rahmani, Hossein Sebe, Nicu Mian, Ajmal The National Engineering Research Center for Robot Visual Perception and Control Technology College of Electrical and Information Engineering the State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body Hunan University Changsha410082 China The School of Architecture and Art Central South University Changsha410082 China The Department of Automation Tsinghua University Beijing100084 China The School of Computing and Communications Lancaster University LA1 4YW United Kingdom The Department of Information Engineering and Computer Science University of Trento Trento38123 Italy The Department of Computer Science The University of Western Australia WA6009 Australia

Object pose estimation is a fundamental computer vision problem with broad applications in augmented reality and robotics. Over the past decade, deep learning models, due to their superior accuracy and robustness, have increasingly supplanted conventional algorithms reliant on engineered point pair features. Nevertheless, several challenges persist in contemporary methods, including their dependency on labeled training data, model compactness, robustness under challenging conditions, and their ability to generalize to novel unseen objects. A recent survey discussing the progress made on different aspects of this area, outstanding challenges, and promising future directions, is missing. To fill this gap, we discuss the recent advances in deep learning-based object pose estimation, covering all three formulations of the problem, i.e., instance-level, category-level, and unseen object pose estimation. Our survey also covers multiple input data modalities, degrees-of-freedom of output poses, object properties, and downstream tasks, providing the readers with a holistic understanding of this field. Additionally, it discusses training paradigms of different domains, inference modes, application areas, evaluation metrics, and benchmark datasets, as well as reports the performance of current state-of-the-art methods on these benchmarks, thereby facilitating the readers in selecting the most suitable method for their application. Finally, the survey identifies key challenges, reviews the prevailing trends along with their pros and cons, and identifies promising directions for future research. We also keep tracing the latest works at Awesome-Object-Pose-Estimation. Copyright © 2024, The Authors. All rights reserved.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Edge-Computing-Driven Autonomous Ubiquitous Internet of Things in Electricity: Architecture and Challenges 3

Edge-Computing-Driven Autonomous Ubiquitous Internet of Thin...

引用

3rd IEEE Conference on Energy Internet and Energy System Integration, EI2 2019

作者： Liu, Dongqi Zeng, Xiangjun Wang, Yaonan Changsha Unviersity of Science and Technology Changsha China National Engineering Laboratory for Robot Visual Perception and Control Technology Changsha China

ISBN: (纸本)9781728131375

The distribution network is developing towards the direction of Internet of Things in Electricity(IoTE). As an emerging technology of the Internet of Things(IoT), edge computing has great application potential in the future IoTE. This paper presents an architecture of edge computing-driven IoTE, whose physical layer is an electrical network containing a high proportion of power electronic devices, and its information layer is a 5G-oriented power communication network containing a large amount of power terminal and mass communication data generated by the power terminal. For the physical layer, the virtual synchronous machine(VSM) algorithm are adopted to control the large number of power electronic components so as to guarantee the dynamic stability of physical power network. For the information layer, a new layered distributed information processing architecture, the Edge Computing Architecture, is introduced to address the needs of information monitoring and protection control in the smart distribution network. The architecture of the edge computing, and the applications that may be deployed in the edge computing gateways are introduced, and the challenges may be faced by the IoTE in the future development are further discussed. © 2019 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Model-free design of stochastic lqr controller from reinforcement learning and primal-dual optimization perspective

arXiv

引用

arXiv 2021年

作者： Li, Man Qin, Jiahu Zheng, Wei Xing Wang, Yaonan Kang, Yu Department of Automation University of Science and Technology of China Hefei230027 China School of Computing Engineering and Mathematics Western Sydney University SydneyNSW2751 Australia College of Electrical and Information Engineering Hunan University Changsha410082 China National Engineering Laboratory for Robot Visual Perception and Control Technology Changsha410082 China

To further understand the underlying mechanism of various reinforcement learning (RL) algorithms and also to better use the optimization theory to make further progress in RL, many researchers begin to revisit the linear-quadratic regulator (LQR) problem, whose setting is simple and yet captures the characteristics of RL. Inspired by this, this work is concerned with the model-free design of stochastic LQR controller for linear systems subject to Gaussian noises, from the perspective of both RL and primal-dual optimization. From the RL perspective, we first develop a new model-free off-policy policy iteration (MF-OPPI) algorithm, in which the sampled data is repeatedly used for updating the policy to alleviate the data-hungry problem to some extent. We then provide a rigorous analysis for algorithm convergence by showing that the involved iterations are equivalent to the iterations in the classical policy iteration (PI) algorithm. From the perspective of optimization, we first reformulate the stochastic LQR problem at hand as a constrained non-convex optimization problem, which is shown to have strong duality. Then, to solve this non-convex optimization problem, we propose a model-based primal-dual (MBPD) algorithm based on the properties of the resulting Karush-Kuhn-Tucker (KKT) conditions. We also give a model-free implementation for the MB-PD algorithm by solving a transformed dual feasibility condition. More importantly, we show that the dual and primal update steps in the MB-PD algorithm can be interpreted as the policy evaluation and policy improvement steps in the PI algorithm, respectively. Finally, we provide one simulation example to show the performance of the proposed algorithms. Copyright © 2021, The Authors. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Automatic basketball detection in sport video based on R-FCN and soft-NMS 2019

Automatic basketball detection in sport video based on R-FCN...

引用

4th International Conference on Automation, control and robotics engineering, CACRE 2019

作者： Liang, Qiaokang Mei, Li Wu, Wanneng Sun, Wei Wang, Yaonan Zhang, Dan Key Laboratory of Intelligent Robot Technology in Electronic Manufacturing National Engineering Laboratory for Robot Vision Perception and Control Hunan University Changsha410082 China Department of Mechanical Engineering York University TorontoM3J 1P3 Canada

ISBN: (纸本)9781450371865

In basketball videos, the ball is always so small in the camera that its appearance feature is hard to be extracted. In this paper, we introduce a deep-learning technology to detect the basketball. Specifically, we train our basketball detection model based on the Region-based Fully Convolutional Networks (R-FCN) which uses the fully convolutional Residual Network (ResNet) as the backbone network. What’s more, we apply some new techniques including Online Hard Example Mining (OHEM), Soft-NMS and multi-scale training strategy to achieve higher detection accuracy. In detail, the OHEM method can reduce the cost of fine-tuning during training by calculating the loss of the RoIs. Soft-NMS can reduce the false positive rate by decreasing the object detection score between the overlap object. And the multi-scale training can improve the detection performance by receiving the good feature from the object with different scale. Finally, we achieve a mean average precision (mAP) value of 89.7% on a public basketball dataset. It proves that applying the deep-learning approach to basketball detection is effective. © 2019 Copyright is held by the owner/author(s).

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

PL-VINS: Real-Time Monocular Visual-Inertial SLAM with Point and Line Features

arXiv

引用

arXiv 2020年

作者： Fu, Qiang Wang, Jialong Yu, Hongshan Ali, Islam Guo, Feng He, Yijia Zhang, Hong National Engineering Laboratory for Robot Visual Perception and Control Technology Hunan University China Department of Computing Science University of Alberta Canada Megvii Technology Beijing China

Leveraging line features to improve localization accuracy of point-based visual-inertial SLAM (VINS) is gaining interest as they provide additional constraints on scene structure. However, real-time performance when incorporating line features in VINS has not been addressed. This paper presents PL-VINS, a real-time and high-efficiency optimization-based monocular VINS method with point and line features, developed based on the state-of-the-art point-based VINS-Mono [1]. We observe that current works use the LSD [2] algorithm to extract line features;however, LSD is designed for scene shape representation instead of the pose estimation problem, which becomes the bottleneck for the real-time performance due to its high computational cost. In this paper, we modify the LSD algorithm by studying a hidden parameter tuning and length rejection strategy. The modified LSD algorithm can run at least three times as fast as LSD. Further, by representing space lines with the Plücker coordinates, the residual error in line estimation is modeled in terms of the point-to-line distance, which is then minimized by iteratively updating the minimum four-parameter orthonormal representation of the Plücker coordinates. Experiments in a public benchmark dataset show that the localization error of our method is 12-16% less than that of VINS-Mono at the same pose update frequency. The source code of our method is available at: https://***/cnqiangfu/PL-VINS. Copyright © 2020, The Authors. All rights reserved.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Online Parameter Estimation For Uncertain robot Manipulators With Fixed-time Convergence

Online Parameter Estimation For Uncertain Robot Manipulators...

引用

IEEE Conference on Industrial Electronics and Applications (ICIEA)

作者： Chengzhi Zhu Yiming Jiang Chenguang Yang Key Laboratory of Autonomous Systems and Networked Control College of Automation Science and Engineering South China University of Technology Guangzhou China National Engineering Laboratory for Robot Visual Perception and Control Hunan University Changsha China Bristol Robotics Laboratory University of the West of England Bristol UK

ISBN: (数字)9781728151694

ISBN: (纸本)9781728151700

For traditional parameter estimation schemes of uncertain robot, most of them were proposed to identify unknown parameter with desired precision, but few of them focused on the convergence time. Recently finite-time estimation techniques have been proposed by scholars to achieve estimation in finite time. In this paper, we proposed a novel estimation scheme for uncertain robot systems with fixed time instead of finite time. In order to avoid using acceleration signals during the estimation, a kind of auxiliary filtering technique was employed. Besides, a continuous and recursive update law was employed for the parameter estimation such that the computational burdens of real-time inversion of square matrices could be avoided. Finally the effectiveness of the identification algorithm is verified based on a 2-DOF uncertain robot model.

关键词： Parameter estimation Filtering Heuristic algorithms Estimation Real-time systems Manipulator dynamics Convergence

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：