检索结果-内蒙古大学图书馆

第32届中国控制与决策会议

作者： Ye Ji Yaping Dai Junjie Ma Kaixin Zhao Yanyan Cheng School of Automation Beijing Institute of Technology State Key Laboratory of Intelligent Control and Decision of Complex System

The research of image deblurring plays an important role in the digital image *** order to reduce image blurring problems,a content constraint loss（CCL） function in the generative adversarial network（GAN）is *** SSIM loss and the perceptual loss constitute the CCL function,which makes the trained generative model *** CCL function as the content constraint loss component and the adversarial loss component constitute the total *** total loss is optimized by the iterative training to further improve the stability of the network model,and the image blurring will be *** the test experiment of the open source image dataset MNIST,CIFAR10/100 and CELEBA,the CCL function is used as the content constraint loss component of the generative adversarial network,the effect of image deblurring has obvious promotion in the structural similarity measure and visual appearance.

关键词： Image deblurring Generative adversarial network Content constraint loss Adversarial loss

来源：评论

学校读者我要写书评

暂无评论

Event-triggered Observability: A Set-membership Perspective

arXiv

引用

arXiv 2022年

作者： Zheng, Kaikai Shi, Dawei Chen, Tongwen State Key Laboratory of Intelligent Control and Decision of Complex Systems School of Automation Beijing Institute of Technology Beijing100081 China Department of Electrical and Computer Engineering University of Alberta EdmontonABT6G 1H9 Canada

This work attempts to discuss the observability of linear time-invariant systems with event-triggered measurements. A new notion of observability, namely, ǫ-observability is defined with parameter ǫ, which relates to the worst-case performance of inferring the initial state based on not only the received measurement but also the implicit information in the event-triggering conditions at no-event instants. A criterion is developed to test the proposed ǫ-observability of discrete-time linear systems, based on which an iterative event-triggered set-membership observer is designed to evaluate a set containing all possible values of the state. The proposed set-membership observer is designed as the outer approximation of the ellipsoids predicted based on previous state estimates and the ellipsoids inferred by fusing the received measurement and communication conditions, which is optimal in the sense of trace at each step and is proved to be asymptotically bounded. The efficiency of the proposed event-triggered set-membership state observer is verified by numerical experiments. © 2022, CC BY.

关键词： state estimation

来源：评论

学校读者我要写书评

暂无评论

Long and Short-Term Constraints Driven Safe Reinforcement Learning for Autonomous Driving

arXiv

引用

arXiv 2024年

作者： Hu, Xuemin Chen, Pan Wen, Yijun Tang, Bo Chen, Long School of Artificial Intelligence Hubei University Wuhan430062 China Key Laboratory of Intelligent Sensing System and Security Hubei University Ministry of Education Hubei Wuhan430062 China Department of Electrical and Computer Engineering Worcester Polytechnic Institute WorcesterMA01609 United States State Key Laboratory of Multimodal Artificial Intelligence Systems State Key Laboratory of Management and Control for Complex Systems Chinese Academy of Sciences Beijing100190 China WAYTOUS Inc. Beijing100083 China Shenzhen518107 China Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi’an710049 China

Reinforcement learning (RL) has been widely used in decision-making and control tasks, but the risk is very high for the agent in the training process due to the requirements of interaction with the environment, which seriously limits its industrial applications such as autonomous driving systems. Safe RL methods are developed to handle this issue by constraining the expected safety violation costs as a training objective, but the occurring probability of an unsafe state is still high, which is unacceptable in autonomous driving tasks. Moreover, these methods are difficult to achieve a balance between the cost and return expectations, which leads to learning performance degradation for the algorithms. In this paper, we propose a novel algorithm based on the long and short-term constraints (LSTC) for safe RL. The short-term constraint aims to enhance the shortterm state safety that the vehicle explores, while the long-term constraint enhances the overall safety of the vehicle throughout the decision-making process, both of which are jointly used to enhance the vehicle safety in the training process. In addition, we develop a safe RL method with dual-constraint optimization based on the Lagrange multiplier to optimize the training process for end-to-end autonomous driving. Comprehensive experiments were conducted on the MetaDrive simulator. Experimental results demonstrate that the proposed method achieves higher safety in continuous state and action tasks, and exhibits higher exploration performance in long-distance decision-making tasks compared with state-of-the-art methods. Copyright © 2024, The Authors. All rights reserved.

关键词： Lagrange multipliers

来源：评论

学校读者我要写书评

暂无评论

Bionic Geomagnetic Navigation for Autonomous Underwater Vehicle with Temporal Attention-based Data-Driven Dead Reckoning

TechRxiv

引用

TechRxiv 2023年

作者： Yang, Songnan Zhang, Xiaohui Liu, Ding Bai, Wenqi Liu, Fan The School of Automation and Information Engineering Xi’an University of Technology Shaanxi 710048 China The Institute of Advanced Navigation and Electromagnetics Xi’an University of Technology Shaanxi 710048 China The State Key Laboratory of Eco-Hydraulics in the Northwest Arid Region of China Xi’an University of Technology Shaanxi 710048 China The Shaanxi Key Laboratory of Complex System Control with the Intelligent Information Processing Xi’an University of Technology Shaanxi 710048 China

Numerous studies have demonstrated that numerous animal species are capable of goal-directed navigation using environmental information for dead reckoning. The stable magnetic field of the earth provides important information for the migration of animals over long distances. Inspired by the goal-directed navigation of animals, a novel Geomagnetic Navigation with Temporal Attention-based Data-Driven Dead Reckoning (Attention-DR) method for Autonomous Underwater Vehicles (AUV) is presented in this article, which only utilizes the inclination angle (I) and the declination angle (D) of the Geomagnetic Field (GF) for underwater navigation without any prior knowledge of the geographical location or geographic map. This article proposes a Temporal Attention-based Long Short-Term Memory (TA-LSTM) neural network by combining history, GF information, and location in time series to achieve the optimization heading angle of underwater dead reckoning. Due to its ability to adjust the utilization weights of local and global temporal information in the path, TA-LSTM possesses the capability to continually update neural network model parameters through a data-driven process during the navigation process. This enables the Attention-DR to achieve efficient navigation in regions without prior magnetic maps and overcome navigation failures caused by magnetic field anomalies. The simulation outcomes confirm the efficiency, precision, and practicability of the proposed algorithm. © 2023, CC BY.

关键词： Navigation

来源：评论

学校读者我要写书评

暂无评论

Prescriptive Manufacturing in Society 5.0: Human Autonomous Organizations and “On-Demand” Smart Services

引用

IFAC-PapersOnLine 2024年第3期58卷 139-144页

作者： Xiao Wang Jun Huang Jing Yang Xiaoxing Wang Fei-Yue Wang Engineering Research Center of Autonomous Unmanned System Technology Ministry of Education Anhui University Hefei 230601 China Macau University of Science and Technology Macau 999078 China School of artificial Intelligence University of Chinese Academy of Sciences Beijing 100049 China Beijing SANBODY Technology Company Ltd Beijing 214000 China Qingdao Academy of Intelligent Industries Qingdao 266109 China State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing 100190 China

Society 5.0 addresses human behaviors and their consequences in social management, while emphasizing the integration of Cyber-Physical-Social Spaces to achieve a highly connected and circulated society. Prompted by the various social needs expressed on social media platforms, Prescriptive Manufacturing (PM) emerges as a proactive pre-manufacturing paradigm. This paper delves into PM within the context of Society 5.0. It highlights the development of Human Autonomous Organizations (HAOs) and the delivery of “on-demand” smart services, which are customized to meet both individual and broader societal demands. The social sensing framework of social needs across different social media platforms that enables the identification and anticipation of consumer demands in real-time is illustrated. This data-driven insight leverages artificial intelligence and robotics to facilitate customized design and on-demand production, aligning product creation with real user requirements. Furthermore, we discuss the implementation of human-centered smart services in PM, such as virtual try-on, virtual fitting, personalized retrofit, user-friendly scheduling, and proximity logistics, which not only enhance consumer experience but also optimize resource efficiency By examining the symbiosis between technological advancements and social need fulfillment, this study sheds light on how PM and HAOs can foster a more responsive, sustainable, and inclusive economy in society, ultimately contributing to a higher quality of life and sustainable development goals of Society 5.0.

关键词： Social Needs Sensing Human Autonomous Organization Customized Design personalized fine-tuning prescriptive manufacturing Human-Centric Smart Services

来源：评论

学校读者我要写书评

暂无评论

Model Free Adaptive control Algorithm based on ReOSELM for Autonomous Driving Vehicles

Model Free Adaptive Control Algorithm based on ReOSELM for A...

引用

第40届中国控制会议

作者： Xiaofei Zhang Hongbin Ma Zhichao Wang Mingyu Fan Bolin Zhao School of Automation Beijing Institute of Technology State Key Laboratory of Intelligent Control and Decision of Complex Systems Beijing Institute of Technology The College of Computer Science and Artificial Intelligence Wenzhou University

Different road conditions and dynamic environment bring significant challenges to the control system of autonomous driving vehicle（ADV）. As is known, historical data collected from ADV contains valuable information about control systems,therefore, it is a promising thing to study adaptive control algorithms that have certain learning ability. In order to improve the control performance of ADV and the efficiency in data usage, in this paper, a model free adaptive control algorithm based on regularized online sequential extreme learning machine（ReOSELM） is introduced, it is difficult to analyze the algorithm based on neural network, and the system stability by improved update algorithm of ReOSELM is proved. Simulation results indicate that the proposed algorithm is effective in improving control precision when ADV is turning, and experimental results on an autonomous driving vehicle show that this algorithm is effective in real environment.

关键词： data-driven control autonomous driving vehicle model free adaptive control regularized online sequential extreme learning machine

来源：评论

学校读者我要写书评

暂无评论

Active Vibration Isolation for A Parallel Wheel-legged Robot Based on Dynamic Model

Research Square

引用

Research Square 2020年

作者： Guo, Fei Wang, Shoukun Yue, Binkai Wang, Junzheng State Key Laboratory of Intelligent Control and Decision of Complex System School of Automation Beijing Institute of Technology Beijing100081 China

来源：评论

学校读者我要写书评

暂无评论

Regional Geomagnetic Map Construction based on Support Vector Machine Residual Kriging

Regional Geomagnetic Map Construction based on Support Vecto...

引用

第三十九届中国控制会议

作者： Tong Liu Xingyu Li Mengyin Fu Zhaoxiang Liang Beijing Institute of Technology Key Laboratory of Complex System Intelligent Control and Decision Nanjing University of Science and Technology

Regional geomagnetic maps are widely used in geomagnetic navigation and magnetic anomaly detection. However,the complexity of geomagnetic spatial trend changes and the spatial sparseness of the geomagnetic data affect the accuracy of regional geomagnetic map construction. In order to improve the accuracy of regional geomagnetic maps, this paper proposes the Support Vector Machine Residual Kriging method（SVMRKriging）. First, Support Vector Machine（SVM） is used to fit the geomagnetic trend changes, then the residual component is interpolated by ordinary Kriging, and finally these two parts are added to construct a regional geomagnetic map. Experiments were performed using geomagnetic grid data and aeromagnetic data. The experiment results show that SVMRKriging method can improve the accuracy of regional geomagnetic maps with geomagnetic trend changes.

关键词： Regional Geomagnetic Map Kriging Support Vector Machine

来源：评论

学校读者我要写书评

暂无评论

Fractional-order Virtual Model control for Trotting Motion of Quadruped Robot

Fractional-order Virtual Model Control for Trotting Motion o...

引用

Chinese control and decision Conference, CCDC

作者： JiangBo Zhao Shicheng Ma Shanshuai Niu Junzheng Wang Key Laboratory of Complex System Intelligent Control and Decision School of Automation Beijing Institute of Technology Beijing China

ISBN: (数字)9781728158556

ISBN: (纸本)9781728158563

Virtual model control (VMC) is the most widely-used method for motion control of quadruped robot. VMC could decrease the abrupt ground reaction force and realize compliant interaction between foot and ground at the cost of control precision. To improve the robustness and precision of VMC, we integrate a fractional-order virtual damper with the classical VMC and propose a fractional-order VMC (FOVMC) for quadruped robot trotting motion. And we provide the implementation details about optimization-based FOVMC parameter tuning method and carry out the trotting experiments on the quadruped robot platform. The experiment results demonstrate that FOVMC could improve the precision of single-leg motion control and achieve a better whole-body trotting performance than classical VMC.

关键词： Legged locomotion Motion control Robot kinematics Foot Torque Force

来源：评论

学校读者我要写书评

暂无评论

High-Flexibility Locomotion and Whole-Torso control for a Wheel-Legged Robot on Challenging Terrain*

High-Flexibility Locomotion and Whole-Torso Control for a Wh...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Kang Xu Shoukun Wang Xiuwen Wang Junzheng Wang Zhihua Chen Daohe Liu Key Laboratory of Drive and Control of Servo Motion System Beijing Institute of Technology Beijing China State Key Laboratory of Intelligent Control and Decision of Complex System Beijing Institute of Technology Beijing China

ISBN: (数字)9781728173955

ISBN: (纸本)9781728173962

In this paper, we propose a parallel six-wheel-legged robot that can traverse irregular terrain while carrying objectives to do heavy-duty work. This robot is equipped with six Stewart platforms as legs and tightly integrates the additional degrees of freedom introduced by the wheels. The presented control strategy with physical system used to adapt the diverse degrees of each leg to irregular terrain such that robot increases the traversability, and simultaneously to maintain the horizontal whole-torso pose. This strategy makes use of Contact Scheduler (CS) and Whole-Torso control (WTC) to control the multiple degrees of freedom (DOF) leg for performing high-flexibility locomotion and adapting the rough terrain like actively parallel suspension system. We conducted experiments on flat, slope, soft and sand-gravel surface, which validate the proposed control method and physical system. Especially, we attempt to traverse over sand-gravel terrain with 3 people about 240kg payload.

关键词： Legged locomotion Wheels Torso Force Kinematics

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：