检索结果-内蒙古大学图书馆

IEEE International Conference on Automation Science and Engineering (CASE)

作者： Ao Zhang Zhen Shen Qihang Fang Jian Yang Gang Xiong Xisong Dong State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences Beijing China Guangdong Engineering Research Center of 3D Printing and Intelligent Manufacturing Cloud Computing Center Chinese Academy of Sciences Dongguan China

ISBN: (数字)9798350358513

ISBN: (纸本)9798350358520

Point clouds have increasingly become the preferred representation for various visual and graphical applications. Denoising diffusion probability models (DDPMs) have demonstrated remarkable applications in areas such as 3D point cloud generation. Furthermore, the ability to generate or reconstruct high-resolution, high-fidelity point clouds stands as a pivotal requirement within this domain. To this end, we propose a geometry aware diffusion model for 3D shape generation. Considering that the global latent shape space overlooks the inherent topology and fine-grained information within the 3D shape itself, we introduce a geometry latent space that operates in a cascaded relationship with the global shape latent space. This latent space combines the global shape latent with the geometry latent space. To generate high-quality point clouds, we conduct training on the geometry latent space, which yields superior results compared to training solely on the global shape latent space. Specifically, we represent the point cloud as a set of unordered point representations with positional embedding. We further employ a geometry aware attention block to model global geometric relationships and retaining local detail features, capturing the inductive bias of the 3D geometric structure of the point cloud. We conduct experiments across various ShapeNet benchmarks, and our approach has demonstrated substantial advancements in the generation of point clouds when compared to the baseline.

关键词： Point cloud compression Geometry Training Solid modeling Visualization Three-dimensional displays Shape Noise reduction Diffusion models Topology

来源：评论

学校读者我要写书评

暂无评论

An Auto-Parallel Method for Deep Learning Models Based on Genetic Algorithm

An Auto-Parallel Method for Deep Learning Models Based on Ge...

引用

International Conference on Parallel and Distributed systems (ICPADS)

作者： Yan Zeng ChengChuang Huang YiJie Ni ChunBao Zhou JiLin Zhang Jue Wang MingYao Zhou MeiTing Xue YunQuan Zhang School of Computer Science and Technology Hangzhou Dianzi University Hangzhou China Key Laboratory for Modeling and Simulation of Complex Systems Ministry of Education Hangzhou China Data Security Governance Zhejiang Engineering Research Center Hangzhou China School of ITMO Joint Institute Hangzhou Dianzi University Hangzhou China Institute of Computer Network Information Center of the Chinese Academy of Sciences Beijing China HuaWei State Key Laboratory of Computer Architecture Institute of Computing Technology of the Chinese Academy of Sciences Beijing China

As the size of datasets and neural network models increases, automatic parallelization methods for models have become a research hotspot in recent years. The existing auto-parallel methods based on machine learning or graph algorithms still have issues with search efficiency and applicability. This paper proposes an automatic parallel method based on a dual-population genetic algorithm, TGA, which transforms model partitioning and placement into an integer linear programming problem and constructs a cost model to evaluate the solution. The solution space is built using the neural network’s dataflow graph and device cluster’s topology, and the dual-population genetic algorithm is used to search for the optimal model parallel strategy. Experiments with various models show that the proposed method can improve single-step execution time by up to 42% compared to the Baechi method and up to 37.7% compared to the Hierarchical method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generating adjacency-constrained subgoals in hierarchical reinforcement learning 20

Generating adjacency-constrained subgoals in hierarchical re...

引用

Proceedings of the 34th International Conference on Neural Information Processing systems

作者： Tianren Zhang Shangqi Guo Tian Tan Xiaolin Hu Feng Chen Department of Automation Tsinghua University Department of Civil and Environmental Engineering Stanford University Department of Computer Science and Technology Tsinghua University and Beijing National Research Center for Information Science and Technology and State Key Laboratory of Intelligent Technology and Systems Department of Automation Tsinghua University and Beijing Innovation Center for Future Chip and LSBDPA Beijing Key Laboratory

ISBN: (纸本)9781713829546

Goal-conditioned hierarchical reinforcement learning (HRL) is a promising approach for scaling up reinforcement learning (RL) techniques. However, it often suffers from training inefficiency as the action space of the high-level, i.e., the goal space, is often large. Searching in a large goal space poses difficulties for both high-level subgoal generation and low-level policy learning. In this paper, we show that this problem can be effectively alleviated by restricting the high-level action space from the whole goal space to a k-step adjacent region of the current state using an adjacency constraint. We theoretically prove that the proposed adjacency constraint preserves the optimal hierarchical policy in deterministic MDPs, and show that this constraint can be practically implemented by training an adjacency network that can discriminate between adjacent and non-adjacent subgoals. Experimental results on discrete and continuous control tasks show that incorporating the adjacency constraint improves the performance of state-of-the-art HRL approaches in both deterministic and stochastic environments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

research on Attention-Based Predictive Model in Multi-Agent Reinforcement Learning

Research on Attention-Based Predictive Model in Multi-Agent ...

引用

Chinese Automation Congress (CAC)

作者： Weixin Hong Huaiyu Wu Zhen Shen Gang Xiong Bin Hu Xisong Dong Fei-Yue Wang The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China The Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences Beijing China The Guangdong Engineering Research Center of 3D Printing and Intelligent Manufacturing Cloud Computing Center Chinese Academy of Sciences Dongguan China

Agent actions planning is a challenging problem in multi-agent reinforcement learning. Recent methods typically build their predictive models by full connection layers, but the shortage of utilizing action and observation information in complex multi-agent environment limits the performance of those methods. To address this issue, we propose a novel predictive model based on self-attention mechanism which can effectively use information in the environment. Specifically, we use the MADDPG algorithm to make a dataset where the input is the observations and actions and the output is the predicted observations and rewards. We also introduce the multi-head attention mechanism into the predictive model so that the predictive model can better combine different information for prediction. We perform extensive experiments to assess the proposed method and compare it with the representative baseline approach. Results demonstrate that our method outperforms competitors with higher accuracy in different tasks. Our implementation will be publicly available.

关键词： Automation Reinforcement learning Predictive models Prediction algorithms Planning Task analysis

来源：评论

学校读者我要写书评

暂无评论

Fast Calibration of Agent-Based Model using Mean-Field Approach

Fast Calibration of Agent-Based Model using Mean-Field Appro...

引用

Digital Twins and Parallel Intelligence (DTPI), IEEE International Conference on

作者： Wanze Lu Gang Xiong Xudong Liu Sheng Liu Peijun Ye Liguo Zhang Faculty of Information and Technology Beijing University of Technology Beijing China The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China The Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences Beijing China The Guangdong Engineering Research Center of 3D Printing and Intelligent Manufacturing The Cloud Computing Center Chinese Academy of Sciences Beijing China

Calibration of simulation system is an important step to establish a reliable simulation system. In this paper, based on the existing mean field calibration theory, a new fast calibration method for complex urban traffic system based on mean field method is discussed by establishing the relationship between the macroscopic observation quantity and the proxy microscopic parameters. This method makes use of the machine learning method, and reduces the dimension of the state vector in the discrete time period, so that the dimension of the macroscopic state transition matrix is lower, and then reduces the computational complexity of the system calibration. Finally, the calibration results of the simulation system were verified by SUMO traffic simulation software.

关键词： Visualization Computational modeling Microscopy Digital twin Machine learning Traffic control Reliability theory

来源：评论

学校读者我要写书评

暂无评论

Parallel systems Framework: A Safe and Efficient Testbed for Unmanned Ships

Parallel Systems Framework: A Safe and Efficient Testbed for...

引用

Digital Twins and Parallel Intelligence (DTPI), IEEE International Conference on

作者： Long Fu Zhen Shen Xisong Dong Yunjun Han Shuguang Yang Gang Xiong Fei-Yue Wang School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China State Key Laboratory of Multimodal Artificial Intelligence Systems Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences Beijing China Guangdong Engineering Center of 3D Printing and Intelligent Manufacturing Cloud Computing Center Chinese Academy of Sciences Dongguan China Schoool of Rail Transportation Shandong Jiaotong University Jinan China State Key Laboratory for Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China

ISBN: (数字)9798350349252

ISBN: (纸本)9798350349269

In this study, unmanned ships test platform based on the parallel systems framework is proposed to improve the test efficiency and accuracy of unmanned ships in complex ocean environment. The parallel intelligence theory and ACP approach, the efficient interaction between virtual and real environment is achieved, which provides more comprehensive and accurate data for unmanned ships testing. Future research will further expand the scope of virtual environment simulation and improve systems accuracy and response speed to a wider range of unmanned ships test requirements.

关键词： Accuracy Oceans Virtual environments Digital twins Marine vehicles Testing

来源：评论

学校读者我要写书评

暂无评论

USV Path Planning Under Marine Environment Simulation Using DWA and Safe Reinforcement Learning

USV Path Planning Under Marine Environment Simulation Using ...

引用

IEEE International Conference on Automation Science and Engineering (CASE)

作者： Tianci Qu Gang Xiong Hub Ali Xisong Dong Yunjun Han Zhen Shen The State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences Beijing China Guangdong Engineering Research Center of 3D printing and Intelligent Manufacturing Cloud Computing Center Chinese Academy of Sciences Dongguan China Qingdao Academy of Intelligent Industries Qingdao China

It is a challenge for USV navigation due to uncertainties and disturbances in the complex marine environment, which may lead to tilting or collision. However, current path planning methods for USVs lack dynamic environment adaptivity and timeliness. We formulate this problem as a Markov Decision Process (MDP) and combine the advantages of Dynamic Window Approach (DWA) and Safe Reinforcement Learning (RL) to solve it. First, the state encoding output by DWA is employed as observation to reduce the state dimension. Second, a safety-ensured reward function is designed to avoid collision areas. Then, the actor-critic network based on RL algorithm is used. Along with the RL training, the parameters in DWA is adaptively adjusted. Simulation experiments with dynamic waves and ocean currents demonstrate the adaptivity of our method, with avg. trajectory length decreased by 10%, avg. time cost decreased by 17%, and avg. speed increased by 18%, compared to the case using only DWA.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Comparative Study of Four Speed Controllers of Brushless DC Motors for Industrial Applications

引用

IFAC-PapersOnLine 2020年第5期53卷 59-64页

作者： Tariku Sinshaw Tamir Gang Xiong Zhen Shen Xiaoyan Gong Sheng Liu Ehtisham Lodhi Li Wan Xisong Dong State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing 100190 China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing 100049 China Guangdong Engineering Research Center of 3D Printing and Intelligent Manufacturing Cloud Computing Center Chinese Academy of Sciences Donggguan 523808 China Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences Beijing 100190 China Ten Dimensions (Guangdong) Technology Co. Ltd. Foshan 528225 China

Direct current (DC) motors are one of the most important kind of motors and are widely used in robotic and industrial applications. Recently, there have been significant efforts to develop direct current (DC) motors in an attempt to control speed of motors. However, conventional controlling approaches perform undesirably in terms of stability and quick response. Therefore, this paper presents a hybrid intelligent controller configuration for optimized speed control of brushless direct current (BLDC) motors in a factory supervisory control data acquisition (SCADA) system. We compare this hybrid intelligent controller with a conventional PID controller, fuzzy logic controller (FLC), and artificial neural network model reference controller (ANNMRC) in MATLAB , and the results show that the hybrid (neuro-fuzzy) controller performs superior in terms of stability, speed trajectory tracking capability, fast response, and simplicity for implementation.

关键词： BLDC motor fuzzy logic controller (FLC) artificial neural network model reference controller (ANNMRC) hybrid (neuro-fuzzy) control

来源：评论

学校读者我要写书评

暂无评论

Public Service Management System of Urban Placenames for Sanya Case

Public Service Management System of Urban Placenames for San...

引用

2016年IEEE/INFORMS服务运筹、物流与信息化国际会议

作者： Wei Liu Fenghua Zhu Xisong Dong Gang Xiong Yongqiang Zhu Huazhou Wu Taozhong Liu Yuexiang Hu The State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of Sciences College of Information Science and Technology Beijing University of Chemical Technology Beijing Engineering Research Center of Intelligent Systems and Technology Institute of AutomationChinese Academy of Sciences Dongguan Research Institute of CASIA Cloud Computing CenterChinese Academy of Sciences Instituteof Smart Education Systems Qingdao Academy of Intelligent Industries Xiamen Kingtopinfo Technology Co.Ltd

The rapidly growing amount of social interaction makes geographical names an important and indispensable medium for people in social and economic exchanges. Especially in today's information age, the frequency of usage of name information has increased greatly. In this paper, a method for the construction of a cities' names database is proposed. This will include the processing of the outcome data from the names pilot survey, the consolidation into the database, the development of the cloud service and management system based on the geographical names and zoning database, and the construction of a relatively complete set of the public service system of geographical names based on a website and touch screen interface. The method has been applied in San Ya, China successfully, and very good results have been obtained. The platform can provide not only accurate and convenient services of geographical names, but also promote its economic development and social progress.

关键词： Public service system geographical names geographical names database cloud service platform

来源：评论

学校读者我要写书评

暂无评论

A process simulation-based quantitative HAZOP analysis method

A process simulation-based quantitative HAZOP analysis metho...

引用

IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI

作者： Jianlong Chen Xiwei Liu Sifeng Jing Gang Xiong Xiuqin Shang Changjian Cheng The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Chemical Engineering and the Environment Beijing Institute of Technology Beijing China Institute of Smart Education Systems Qingdao Academy of Intelligent Industries Qingdao China Cloud Computing Center Chinese Academy of Sciences Dongguan China Beijing Engineering Research Center for Intelligent Systems and Technology Chinese Academy of Science Beijing China

ISBN: (纸本)9781509029280

Hazard and operability (HAZOP) methodology is an important Process Hazard Analysis (PHA) technique for ensuring safety production in chemical industry. Due to the lacks of accuracy and reliability in manual HAZOP analysis, the quantitative HAZOP analysis has gained more and more attention. In this work, a process simulation-based HAZOP analysis method and its flow chart are proposed. And the proposed method is applied to a steam generation subsystem in an ethylene pyrolysis unit on a dynamic process simulation platform. All deviation parameters of system can be calculated and described quantitatively. The case study shows the proposed method is useful for the effective risk prevention and control measures in a chemical process.

关键词： Analytical models Hazards Manuals Process control Boilers

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：