检索结果-内蒙古大学图书馆

Policy iteration based Q-learning for linear nonzero-sum quadratic differential games

Science China(Information Sciences) 2019年第5期62卷 195-213页

作者： Xinxing LI Zhihong PENG Li LIANG Wenzhong ZHA School of Automation Beijing Institute of Technology State Key Laboratory of Intelligent Control and Decision of Complex System Information Science Academy China Electronics Technology Group Corporation

In this paper, a policy iteration-based Q-learning algorithm is proposed to solve infinite horizon linear nonzero-sum quadratic differential games with completely unknown dynamics. The Q-learning algorithm, which employs off-policy reinforcement learning(RL), can learn the Nash equilibrium and the corresponding value functions online, using the data sets generated by behavior policies. First, we prove equivalence between the proposed off-policy Q-learning algorithm and an offline PI algorithm by selecting specific initially admissible polices that can be learned online. Then, the convergence of the off-policy Qlearning algorithm is proved under a mild rank condition that can be easily met by injecting appropriate probing noises into behavior policies. The generated data sets can be repeatedly used during the learning process, which is computationally effective. The simulation results demonstrate the effectiveness of the proposed Q-learning algorithm.

关键词： adaptive dynamic programming ADP Q-learning reinforcement learning RL linear nonzerosum quadratic differential games policy iteration PI off-policy

来源：评论

学校读者我要写书评

暂无评论

Effective Throughput Optimization of SAG Milling Process Based on BPNN and Genetic Algorithm

Effective Throughput Optimization of SAG Milling Process Bas...

引用

Industrial Cyber-Physical systems (ICPS)

作者： Zhenhong Liao Ce Xu Wen Chen Qifu Chen Feng Wang Jinhua She School of Automation China University of Geosciences Wuhan China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan China Department of Mineral Resources Development and Utilization Changsha Research Institute of Mining and Metallurgy CO. LTD. Chagnsha China Hunan Valin Lianyuan Iron And Steel Co. Ltd. Loudi China School of Engineering Tokyo University of Technology Hachioji Tokyo Japan

Grinding is an energy-consuming process in mineral processing industry. Improving grinding processing capacity per unit power consumption is an effective means to reduce grinding production cost. In this paper, a new index for evaluating the effective processing throughput of SAG milling is proposed. The production process model is established by BP neural network (BPNN). Through combining the process mechanism and production constraints, the genetic algorithm is adopted to optimize the operating parameters of the SAG milling process to maximize the effective throughput, thus improving the grinding efficiency. The experimental results showed that through optimization of effective throughput function proposed in this paper, the SAG mill processing capacity has been increased by 4% and the operating power drawn reduced by 1.12%. It has important guiding significance for the actual production process.

关键词： Industries Power demand Neural networks Milling Throughput Minerals Materials processing

来源：评论

学校读者我要写书评

暂无评论

Fast ADMM algorithm with efficient communication for distributed optimization of multi-agent systems

Fast ADMM algorithm with efficient communication for distrib...

引用

Chinese automation Congress (CAC)

作者： Shuang Cao Xiaowei Jiang Ming Chi Xianhe Zhang School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation of Complex Systems Engineering Research Center of Intelligent Geodetection Technology Ministry of Education Wuhan China School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China School of Electrical Engineering and Automation Hubei Normal University Huangshi China

ISBN: (纸本)9781665426480

This paper proposes a distributed optimization algorithm based on alternating direction method of multipliers (ADMM) for the distributed optimization problem of multi-agent systems, called ADMM with adaptive penalty terms and conditional communication (for short, AAPCC), by using the local information of agents to update the adaptive penalty terms to speed up the convergence speed while by restricting the communication conditions of agents to reduce the communication cost of agents in the iterative process. Thus, the communication cost of the system is reduced while the convergence speed is ensured. Numerical simulation experiments show that in a multi-agent system, compared with the classical ADMM, AAPCC can better balance the convergence speed and communication cost, suitable for multi-agent systems that require fast convergence speed and less communication cost.

关键词： Costs automation Numerical simulation Convex functions Stability analysis Iterative algorithms Optimization

来源：评论

学校读者我要写书评

暂无评论

Self-Attention based Temporal Intrinsic Reward for Reinforcement Learning

Self-Attention based Temporal Intrinsic Reward for Reinforce...

引用

Chinese automation Congress (CAC)

作者： Zhuo Jiang Daiying Tian Qingkai Yang Zhihong Peng School of Automation Beijing Institute of Technology Beijing China State Key Laboratory of Intelligent Control and Decision of Complex System Beijing China Peng Cheng Laboratory Shenzhen China

ISBN: (纸本)9781665426480

This paper proposes a self-attention based temporal intrinsic reward model for reinforcement learning (RL), to synthesize the control policy for the agent constrained by the sparse reward in partially observable environments. This approach can solve the problem of temporal credit assignment to some extent and deal with the low efficiency of exploration. We first introduce a sequence-based self-attention mechanism to generate the temporary features, which can effectively capture the temporal property of the task for the agent. During the training process, the temporary features are employed in each sampled episode to elaborate the intrinsic rewards, which is combined with the extrinsic reward to help the agent learn a feasible policy. Then we use the meta-gradient to update this intrinsic reward model in order that the agent can achieve better performance. Experiments are given to demonstrate the superiority of the proposed method.

关键词： Training automation Reinforcement learning Task analysis

来源：评论

学校读者我要写书评

暂无评论

A dulcimer keys positioning method based on improved Roberts operator

A dulcimer keys positioning method based on improved Roberts...

引用

Chinese automation Congress (CAC)

作者： De Tang Ziying Zhang Xin Chen Zhe Xiao Mengxi Qin School of Automation China University of Geosciences Wuhan China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Ministry of Education Engineering Research Center of Intelligent Technology for Geo-Exploration School of Arts and Communication China University of Geosciences Wuhan China

ISBN: (纸本)9781665426480

It is important for the dulcimer robot to obtain the spatial coordinates of the dulcimer phonemes. However, the traditional manual positioning method is both inefficient and does not meet the intelligence requirements of modern robot. A visual positioning method based on improved Roberts operator is proposed in this paper on account of solving these problems, which can obtain the 3D coordinates of the dulcimer keys in a short time. First, an improved Roberts edge detection operator is designed, which fully detects the edge information of the target by transforming the scale. Further, this paper proposes a dulcimer keys positioning method by fusing RGB and depth image information. This method has a better positioning effect for dulcimer keys with complex background and small size. The experimental results show that the visual positioning method proposed in this paper can effectively identify and locate the dulcimer keys, and it has better robustness to the lighting environment.

关键词： Visualization Three-dimensional displays Robot kinematics Image edge detection Measurement uncertainty Lighting Manuals

来源：评论

学校读者我要写书评

暂无评论

FFPA-Net: Efficient Feature Fusion with Projection Awareness for 3D Object Detection

arXiv

引用

arXiv 2022年

作者： Jiang, Chaokang Wang, Guangming Wu, Jinxing Miao, Yanzi Wang, Hesheng Engineering Research Center of Intelligent Control for Underground Space Ministry of Education School of Information and Control Engineering Advanced Robotics Research Center China University of Mining and Technology Xuzhou221116 China Department of Automation Key Laboratory of System Control and Information Processing of Ministry of Education Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education Shanghai Engineering Research Center of Intelligent Control and Management Shanghai Jiao Tong University Shanghai200240 China The Department of Engineering Mechanics Shanghai Jiao Tong University Shanghai200240 China

Promising complementarity exists between the texture features of color images and the geometric information of LiDAR point clouds. However, there still present many challenges for efficient and robust feature fusion in the field of 3D object detection. In this paper, first, unstructured 3D point clouds are filled in the 2D plane and 3D point cloud features are extracted faster using projection-aware convolution layers. Further, the corresponding indexes between different sensor signals are established in advance in the data preprocessing, which enables faster cross-modal feature fusion. To address LiDAR points and image pixels misalignment problems, two new plug- and-play fusion modules, LiCamFuse and BiLiCamFuse, are proposed. In LiCamFuse, soft query weights with perceiving the Euclidean distance of bimodal features are proposed. In BiLiCamFuse, the fusion module with dual attention is proposed to deeply correlate the geometric and textural features of the scene. The quantitative results on the KITTI dataset demonstrate that the proposed method achieves better feature-level fusion. In addition, the proposed network shows a shorter running time compared to existing methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Object detection

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

Cross-Membrane Penetration to Nucleus of Adherent Cells Using Micropipettes Made by Borosilicate Glass and Quartz 17

Cross-Membrane Penetration to Nucleus of Adherent Cells Usin...

引用

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

作者： Xu, Xinyue Liu, Xiaoming Lin, Yuqing Li, Pengyun Liu, Fengyu Tang, Xiaoqing Huang, Qiang Arai, Tatsuo Beijing Institute of Technology Key Laboratory of Biomimetic Robots and Systems Ministry of Education State Key Laboratory of Intelligent Control and Decision of Complex System Beijing Advanced Innovation Center for Intelligent Robots and Systems School of Mechatronical Engineering Beijing100081 China

ISBN: (纸本)9781728164151

In the biomicromanipulation field, single-cell manipulation is receiving attention due to its potential applications in many aspects of biomedicine. Cell injection, as a typical cell micromanipulation technique, is devoted to introducing foreign substances into living cells by penetrating a fine micropipette into cell membranes. Direct access to the intracellular environment, especially the nucleus, provides an ideal way to gain insight into the dynamic processes of living cells. This paper presents the comparison of controlled penetrations of two types of adherent cells and their nuclei by two types of micropipettes, made of borosilicate glass and quartz respectively. Due to the differences in the cell properties (e.g., height, adhesion strength), two corresponding penetration strategies optimizing the puncture angle and puncture height, are proposed to improve the success rate of the puncture process and the survival rate of cells after the puncture. Real nucleus puncture experiments adopting the optimal puncture strategy for each type of cells and carried out by a semi-automatic system, indicated a better puncture effect by quartz pipettes. Through this method, functional materials such as peptides and nanoparticles can be potentially introduced to perturb the cell activities and then realize its full automation. © 2020 IEEE.

关键词： Quartz

来源：评论

学校读者我要写书评

暂无评论

A threshold segmentation method for non-uniform illumination image based on brightness equalization 2

A threshold segmentation method for non-uniform illumination...

引用

2019 2nd International Conference on advanced Materials, intelligent Manufacturing and automation, AMIMA 2019

作者： Guo, Hui Chen, Peng Huang, Shun Hu, Cong Zhang, Lijun School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex System Wuhan430074 China

In the process of image acquisition, non-uniform illumination images are common due to poor lighting, surface reflection, or a combination of these two factors. In order to improve the quality of image segmentation, an image threshold segmentation method is proposed for non-uniform illumination images. First, the brightness of different regions in the image is compensated to make the brightness background of the whole image consistent. Second, the gradation histogram of the processed image is obtained and simulated into Gaussian distribution curve, then the inflection points of the curve are calculated. Finally, the two inflection points are set as thresholds to segment the image after brightness equalization. This method eliminates the influence of non-uniform illumination to a certain extent and gets a better segmentation effect. Experiments were carried out on images containing several different types of non-uniform illumination. The results demonstrate that the proposed method outperforms the compared enhancement algorithms in threshold segmentation. © Published under licence by IOP Publishing Ltd.

关键词： Luminance

来源：评论

学校读者我要写书评

暂无评论

Image retrieval based on convolutional neural network and linear discriminate analysis 2

Image retrieval based on convolutional neural network and li...

引用

2019 2nd International Conference on advanced Materials, intelligent Manufacturing and automation, AMIMA 2019

作者： Xia, Wei Hu, Cong Cheng, Xianghui Chen, Peng Jin, Xing School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex System Wuhan430074 China

Image retrieval is a hot research topic in the field of computer vision image processing, and the user queries the image database for similar images and produces a list of recommendations. The paper firstly sets forth the research status of image retrieval, then the convolution neural network is briefly introduced. Due to the traditional image retrieval and recommendation system use manual extraction of image features is relatively cumbersome, and the retrieval accuracy is not high research status, the paper proposes an image retrieval method based on the improved convolutional neural network and linear discriminate analysis. Caltech256 and CIFAR-10 datasets were trained using the model in this paper, experimental, results show that the proposed method can effectively improve the performance of retrieval. © Published under licence by IOP Publishing Ltd.

关键词： Search engines

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：