检索结果-内蒙古大学图书馆

3rd International Seminar on Artificial Intelligence, Networking, and Information technology, AINIT 2022

作者： Zhang, Yasong Li, Yue Wang, Xiaoling National Key Laboratory of Science and Technology on Aerospace Intelligence Control Beijing China Beijing Aerospace Automatic Control Institute Beijing China

ISBN: (数字)9781510662971

ISBN: (纸本)9781510662964

The deployments of deep learning models must be highly optimized by experts or hardware suppliers before being used in practice, and it has always been a long-term goal for the compiler community to enable compilers to automatically optimize code. However, there is no feasible solution in practice as running a program costs a considerable amount of optimization time to obtain a desired latency. Aiming at making up for the deficiency of long optimization time of TVM compiler, a novel optimized hybrid aging evolutionary algorithm is proposed to predict the running time of the code and accelerate automatic code optimization for Ansor, an auto-tuning framework for TVM. The algorithm alternately removes the worst and oldest individuals in the population during the evolution process. Unlike previous evolutionary algorithm, if an individual seeks to survive in the evolving population for a long time, it must have excellent scalability and flexibility, not just the individual's own adaptability. In this way, this algorithm not only ensures a strong search capability, but also improves the convergence speed and accuracy, significantly reducing the optimization time of tensor programs for deep learning inference. Experimental results show that the algorithm can accelerate convergence speed. For the same task, our algorithm provides 9% to 16% shorter optimization time on average while achieving similar or better optimization quality (i.e., inference time). © 2023 SPIE.

关键词： Evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

Network Generating Network for Multi-scale Image Classification 2

Network Generating Network for Multi-scale Image Classificat...

引用

2nd Guangdong-Hong Kong-Macao Greater Bay Area Artificial Intelligence and Big Data Forum, AIBDF 2022

作者： Dong, Hang Xiao, Liping Cong, Longjian Zhou, Bin National Key Laboratory of Science and Technology on Aerospace Intelligence Control Beijing China Beijing Aerospace Automatic Control Institute Beijing China

ISBN: (纸本)9781510663107

Features extracted by the neural network do not have scale invariance, which makes multi-scale image recognition and classification a difficult problem. Recent studies have proposed many new ways to solve this problem, such as feature fusion, sensor field transformation, etc. However, none of them essentially solve the problem that the neural network does not have scale invariance. In this paper, we propose a network generating network (NGN) architecture and design the NGNResNet network, which is an improved version of the ResNet network. The network can identify images at three scales simultaneously and has scale invariance. The experimental results show that the NGN structure helps us to improve the classification accuracy of small-scale images by about 10 percentage points, and helps to improve the performance of the network in the face of small targets. © 2023 SPIE.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

Residual stress measurement and analysis of siliceous slate-containing quartz veins

引用

International Journal of Minerals,Metallurgy and Materials 2023年第12期30卷 2310-2320页

作者： Tao Wang Weiwei Ye Yemeng Tong Naisheng Jiang Liyuan Liu School of Civil and Resources Engineering University of Science and Technology BeijingBeijing 100083China School of Materials Science and Engineering University of Science and Technology BeijingBeijing 100083China State Key Laboratory of Strata Intelligent Control and Green Mining Co-founded by Shandong Province and the Ministry of Science and Technology Shandong University of Science and TechnologyQingdao 266590China

Engineering geological disasters such as rockburst have always been a critical factor affecting the safety of coal mine ***,residual stress is considered a feasible method to explain these geomechanical *** this study,electron backscatter diffraction(EBSD)and optical microscopy were used to characterize the rock microcosm.A measuring area that met the requirements of X-ray diffraction(XRD)residual stress measurement was determined to account for the mechanism of rock residual ***,the residual stress of a siliceous slate-containing quartz vein was measured and calculated using the sin^(2) ϕ method equipped with an X-ray *** of microscopic test results showed homogeneous areas with small particles within the millimeter range,meeting the requirements of XRD stress measurement *** was determined as the calibration mineral for slate samples containing quartz *** diffraction patterns of the(324)crystal plane were obtained under different ϕ and φ.The deviation direction of the diffraction peaks was consistent,indicating that the sample tested had residual *** addition,the principal residual stress within the quartz vein measured by XRD was compressive,ranging from 10 to 33 *** maximum principal stress was parallel to the vein trend,whereas the minimum principal stress was perpendicular to the vein ***,the content of the low-angle boundary and twin boundary in the quartz veins was relatively high,which enhances the resistance of the rock mass to deformation and promotes the easy formation of strain concentrations,thereby resulting in residual *** proposed method for measuring residual stress can serve as a reference for subsequent observation and related research on residual stress in different types of rocks.

关键词： residual stress siliceous sand rock microstructure X-ray diffraction electron backscatter diffraction

来源：评论

学校读者我要写书评

暂无评论

Improving aircraft nose wheel steering stability based on control continuation method

引用

science China(Technological sciences) 2025年第3期68卷 258-277页

作者： Qiaozhi YIN Dexu KONG Jiawei WANG Hong NIE Xiaohui WEI National Key Laboratory of Helicopter Aeromechanics Nanjing210016China State Key Laboratory of Mechanics and Control for Aerospace Structures Nanjing210016China Key Laboratory of Fundamental Science for National Defense-Advanced Design Technology of Flight Vehicle Nanjing210016China AVIC Nanjing Engineering Institute of Aircraft Systems Nanjing211106China

Aircraft steering instability during the taxiing process sometimes occurs resulting from the unstable bifurcation phenomena of the nonlinear steering dynamic system,which may lead to severe *** this study,the control parameter design of an aircraft nose wheel steering control system based on the control continuation method is proposed to reduce the number of unstable bifurcation points and enlarge the system stable region area in a certain parameter ***,the aircraft turning performance can be improved ***,a nonlinear aircraft rollout dynamic model is constructed,and the aircraft ground steering bifurcation characteristics are obtained under the open-loop steering control ***,the bifurcation performances under three closed-loop nose wheel steering controllers based on the feedback control method are compared,and the control parameter of the quadratic nonlinear controller with a better bifurcation control effect is selected as a second bifurcation parameter to analyze the directional system stability via two-parameter bifurcation ***,the influence of a crosswind on the aircraft ground steering bifurcation performance is studied for various control parameter ***,the control effects of a traditional proportional-integral-derivative(PID)nose wheel steering control law and the designed feedback controller are compared and ***,a semi-physical experiment of the aircraft nose wheel steering system is carried out,and different control parameter values are selected for test verification,which can further verify the control effect of the designed nonlinear feedback *** results indicate that the feedback control method is unable to vary the system bifurcation types,but appropriate control parameter design can effectively shrink the system instability region and even eliminate the unstable Hopf bifurcation points under some specific working conditions,improving the aircraft ground turning stability re

关键词： aircraft taxiing dynamics and stability nose wheel steering control system control continuation steering bifurcation characteristics quadratic feedback control

来源：评论

学校读者我要写书评

暂无评论

Formal management-specifying approach for model-based safety assessment

引用

Journal of Systems Engineering and Electronics 2023年第6期34卷 1589-1601页

作者： XU Changyi DUAN Yiman ZHANG Chao School of Control Science and Engineering Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of EducationDalian University of TechnologyDalian 116024China State Key Laboratory of Fluid Power and Mechatronic Systems School of Mechanical EngineeringZhejiang UniversityHangzhou 310027China

In the field of model-based system assessment,mathematical models are used to interpret the system ***,the industrial systems in this intelligent era will be more *** management operations will be dynamically set,and the system will be no longer static as it is initially ***,the static model generated by the traditional model-based safety assessment(MBSA)approach cannot be used to accurately assess the *** mainly exists three ***:huge and complex behaviors make the modeling to be trivial manual;Dynamic:though there are thousands of states and transitions,the previous model must be resubmitted to assess whenever new management arrives;Unreusable:as for different systems,the model must be resubmitted by reconsidering both the management and the system itself at the same time though the management is the *** by solving the above problems,this research studies a formal management specifying approach with the advantages of agility modeling,dynamic modeling,and specification design that can be ***,three typical managements are specified in a series-parallel system as a demonstration to show the potential.

关键词： model-based safety assessment(MBSA) management availability reliability maintainability continuous time Markov chain.

来源：评论

学校读者我要写书评

暂无评论

Cooperative Game Theory Based Coordinated Scheduling of Two-Machine Flow-Shop and Transportation

引用

Journal of Systems science & Complexity 2023年第6期36卷 2415-2433页

作者： SUN Wenjuan GONG Hua LIU Peng School of Management Shenyang University of TechnologyShenyang110870China Liaoning Key Laboratory of Intelligent Optimization and Control for Ordnance Industry School of ScienceShenyang Ligong UniversityShenyang110159China

A cooperative game theoretical approach is taken to production and transportation coordinated scheduling problems of two-machine flow-shop(TFS-PTCS problems)with an interstage *** authors assume that there is an initial scheduling order for processing jobs on the *** cooperative sequencing game models associated with TFS-PTCS problems are established with jobs as players and the maximal cost savings of a coalition as its *** properties of cooperative games under two different types of admissible rearrangements are *** TFS-PTCS problems with identical processing time,it is proved that,the corresponding games areσ_(0)-component additive and convex under one admissible *** Shapley value gives a core allocation,and is provided in a computable *** the other admissible rearrangement,the games neither need to beσ_(0)-component additive nor convex,and an allocation rule of modified Shapley value is *** properties of the cooperative games are analysed by a counterexample for general problems.

关键词： Cooperative sequencing games core allocation production and transportation coordinated scheduling Shapley value two-machine flow-shop

来源：评论

学校读者我要写书评

暂无评论

Semisupervised Transfer Boosting (SS-TrBoosting)

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第7期5卷 3431-3444页

作者： Deng, Lingfei Zhao, Changming Du, Zhenbang Xia, Kun Wu, Dongrui Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan 430074 China and also with Shenzhen Huazhong University of Science and Technology Research Institute Shenzhen 518063 China

Semisupervised domain adaptation (SSDA) aims at training a high-performance model for a target domain using few labeled target data, many unlabeled target data, and plenty of auxiliary data from a source domain. Previous works in SSDA mainly focused on learning transferable representations across domains. However, it is difficult to find a feature space where the source and target domains share the same conditional probability distribution. Additionally, there is no flexible and effective strategy extending existing unsupervised domain adaptation (UDA) approaches to SSDA settings. In order to solve the above two challenges, we propose a novel fine-tuning framework, semisupervised transfer boosting (SS-TrBoosting). Given a well-trained deep learning-based UDA or SSDA model, we use it as the initial model, generate additional base learners by boosting, and then use all of them as an ensemble. More specifically, half of the base learners are generated by supervised domain adaptation, and half by semisupervised learning. Furthermore, for more efficient data transmission and better data privacy protection, we propose a source data generation approach to extend SS-TrBoosting to semisupervised source-free domain adaptation (SS-SFDA). Extensive experiments showed that SS-TrBoosting can be applied to a variety of existing UDA, SSDA, and SFDA approaches to further improve their performance. © 2020 IEEE.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

control of 2-D Semi-Markov Jump Systems:A View from Mode Generation Mechanism

引用

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 258-260页

作者： Yunzhe Men Jian Sun Jie Chen the National Key Laboratory of Autonomous Intelligent Unmanned Systems the School of AutomationBeijing Institute of TechnologyBeijing 100081 the Beijing Institute of Technology Chongqing Innovation Center Chongqing 401120China the Department of Control Science and Engineering Tongji UniversityShanghai 201804 the National Key Laboratory of Autonomous Intelligent Unmanned Systems Beijing Institute of TechnologyBeijing 100081China

Dear Editor,Two-dimensional(2-D) systems have wide applications in image data processing,gas absorption and fluid dynamics analysis [1]-[3].When there exist abrupt changes in 2-D systems,they are usually modeled by 2-D Markov jump systems(MJSs) or 2-D semi-Markov jump systems(SMJSs).This letter investigates the control of 2-D SMJSs based on a novel mode generation mechanism,which could avoid mode ambiguousness phenomenon caused by the evolution of system mode in two different *** criterion that guarantees the almost surely exponential stability of the system is obtained.A thermal process is studied to demonstrate the availability of the proposed method.

关键词： Markov mode system

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning of Space Robotic Manipulation with Multiple Safety Constraints 41

Reinforcement Learning of Space Robotic Manipulation with Mu...

引用

第41届中国控制会议

作者： Linfeng Li Yongchun Xie Yong Wang Ao Chen Beijing Institute of Control Engineering Science and Technology on Space Intelligent Control Laboratory

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

To execute numerous space robotic manipulation tasks(for example on-orbit capture,refueling,assembly,etc.) reliably in presence of high environment uncertainty and multi-source obstacles,it is critical to incorporate safety design into trajectory ***-based planning approaches are commxonly used in the literature,yet suffer insufficient robustness to model *** this paper,we study the sample-based,model-free learning of safe space robotic *** consider numerous practical on-orbit working conditions and propose a novel algorithm called Safe policy optimization with multiple constraints(SPOMC).It incorporates two constraints(namely energy constraint and geometric constraint) into the constrained policy optimization framework,maximizing expected total reward while maintaining a bounded expected total *** simulation on rigid robotic arm manipulation at randomized base positions shows the effectiveness of SPOMC.

关键词： Safety Space robotic manipulation Sample-based trajectory optimization Reinforcement learning Constrained Markov decision processes

来源：评论

学校读者我要写书评

暂无评论

A Low Power Method for Multi-FPGA Cascade Loading 3

A Low Power Method for Multi-FPGA Cascade Loading

引用

3rd International Conference on Frontiers of Electronics, Information and Computation Technologies, ICFEICT 2023

作者： Zhang, Ying Zhao, Qian Jiang, Chongwu Wang, Shihui Tao, Yizhu Beijing Aerospace Automatic Control Institute National Key Laboratory of Science and Technology on Aerospace Intelligent Beijing100854 China Yuying Education Group Yuying School Beijing100018 China

ISBN: (纸本)9798350302356

Multi-FPGA with the cascade structure is widely used, which urgently needs the low power loading fuction in the long term working stage. For the same size configuration file, choosing different configuration modes, the configuration circuit design and configuration time could be very different. To analyse the effects for the Multi-FPGA structures, the typical model is researched respectively on master serial (MS), slave serial (SS), slave parallel (SP), active parallel (AP), master parallel (MP), fast SPI, standard SPI and JTAG configuration. And it is demonstrated on the management design of multi-FPGA cascade loading, including factors to be considered, when choosing a configuration mode with overall setup, speed, cost, program upgrade complexity, and applications. Among them, to lower the power, even though the JTAG cascade mode has fast loading speed, the product still needs to be equipped with peripheral devices, such as emulators when upgrading, which is inconvenient to use;the MCU loading mode has slow loading speed, complex overall settings and low cost, and requires a remote computer and network to backup the upgrading;the flash loading mode with fast loading speed, complex overall setting and high cost, supports both remote and local upgrade modes. With all these setting to get the lowest power consumption, the specified type for the MultiFPGA simulation with the cascade structure can only take 2.7 s to load, with the33 MHz clock. © 2023 IEEE.

关键词： Electric power utilization

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：