检索结果-南通市图书馆

Mastering air combat game with deep reinforcement learning

Defence Technology（防务技术） 2024年第4期34卷 295-312页

作者： Jingyu zhu Minchi Kuang Wenqing Zhou Heng Shi jihong zhu Xu Han Department of Precision Instruments Tsinghua UniversityBeijing 100084China Department of Computer Science and Technology Tsinghua UniversityBeijing 100084China Chengdu Aircraft Design&Research Institute Aviation Industry Corporation of ChinaChengdu 610000China

Reinforcement learning has been applied to air combat problems in recent years,and the idea of curriculum learning is often used for reinforcement learning,but traditional curriculum learning suffers from the problem of plasticity loss in neural *** loss is the difficulty of learning new knowledge after the network has *** this end,we propose a motivational curriculum learning distributed proximal policy optimization(MCLDPPO)algorithm,through which trained agents can significantly outperform the predictive game tree and mainstream reinforcement learning *** motivational curriculum learning is designed to help the agent gradually improve its combat ability by observing the agent's unsatisfactory performance and providing appropriate rewards as a ***,a complete tactical maneuver is encapsulated based on the existing air combat knowledge,and through the flexible use of these maneuvers,some tactics beyond human knowledge can be *** addition,we designed an interruption mechanism for the agent to increase the frequency of decisionmaking when the agent faces an *** the number of threats received by the agent changes,the current action is interrupted in order to reacquire observations and make decisions *** the interruption mechanism can significantly improve the performance of the *** simulate actual air combat better,we use digital twin technology to simulate real air battles and propose a parallel battlefield mechanism that can run multiple simulation environments simultaneously,effectively improving data *** experimental results demonstrate that the agent can fully utilize the situational information to make reasonable decisions and provide tactical adaptation in the air combat,verifying the effectiveness of the algorithmic framework proposed in this paper.

关键词： Air combat MCLDPPO Interruption mechanism Digital twin Distributed system

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Fiber Bundle Topology Optimization for Surface Flows

引用

Chinese Journal of Mechanical Engineering 2024年第3期37卷 236-264页

作者： Yongbo Deng Weihong Zhang jihong zhu Yingjie Xu Jan G Korvink State Key Laboratory of Applied Optics Changchun Institute of OpticsFine Mechanics and PhysicsChinese Academy of SciencesChangchun 130033China State IJR Center of Aerospace Design and Additive Manufacturing Northwestern Polytechnical UniversityXi’an 710072China Institute of Microstructure Technology(IMT) Karlsruhe Institute of Technology(KIT)Hermann-von-Helmholtzplatz 176344 Eggenstein-LeopoldshafenGermany

This paper presents a topology optimization approach for the surface flows on variable design *** this approach,the matching between the pattern of a surface flow and the 2-manifold used to define the pattern can be optimized,where the 2-manifold is implicitly defined on another fixed 2-manifold named as the base *** fiber bundle topology optimization approach is developed based on the description of the topological structure of the surface flow by using the differential geometry concept of the fiber *** material distribution method is used to achieve the evolution of the pattern of the surface *** evolution of the implicit 2-manifold is realized via a homeomorphous *** design variable of the pattern of the surface flow and that of the implicit 2-manifold are regularized by two sequentially implemented surface-PDE *** two surface-PDE filters are coupled,because they are defined on the implicit 2-manifold and base manifold,*** surface Navier-Stokes equations,defined on the implicit 2-manifold,are used to describe the surface *** fiber bundle topology optimization problem is analyzed using the continuous adjoint method implemented on the first-order Sobolev *** numerical examples have been provided to demonstrate this approach,where the combination of the viscous dissipation and pressure drop is used as the design objective.

关键词： Fiber bundle Topology optimization 2-manifold Surface flow Material distribution method Porous medium model

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Topology optimization of modular structures with multiple assemblies and applications to airborne shelves

引用

Chinese Journal of Aeronautics 2024年第4期37卷 321-332页

作者： Jie WANG Tong GAO Ming LI jihong zhu Longlong SONG Weihong ZHANG State IJR Center of Aerospace Design and Additive Manufacturing Northwestern Polytechnical UniversityXi’an 710072China Beijing Specialized Machinery Institute Beijing 100143China

This work is devoted to the aeronautical application of topology optimization for modular structures with multiple assemblies that consist of repeated standard modules and optional *** kinds of structures are widely used owing to their transportability,reconfigurability,low manufacturing and service *** this work,the design of airborne shelves with modular structures characterized by the standard module configuration is formulated for the first time as a topology optimization problem of multiple assemblies and multiple load cases subjected to the volume *** is shown that the weighted compliance design of multiple assemblies is a compromising solution compared to the optimization result of each individual assembly of standard ***,the performance of optimized airborne shelves with the modular structures can effectively be ameliorated with the help of reinforcements.

关键词： Topology optimization Modular structures Multiple assemblies Standard modules Airborne shelves

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Study of the Ballistic Impact Behavior of Protective Multi-Layer Composite Armor

引用

Computer Modeling in Engineering & Sciences 2024年第7期140卷 171-199页

作者： Dongsheng Jia Yingjie Xu Liangdi Wang jihong zhu Weihong Zhang State IJR Center of Aerospace Design and Additive Manufacturing Northwestern Polytechnical UniversityXi’an710072China

The abalone shell,a composite material whose cross-section is composed of inorganic and organic layers,has high strength and *** by the abalone shell,several multi-layer composite plates with different layer sequences and thicknesses are studied as bullet-proof material in this *** investigate the ballistic performance of this multi-layer structure,the complete characterization model and related material parameters of large deformation,failure and fracture ofAl_(2)O_(3)ceramics andCarbon Fiber Reinforced Polymer(CFRP)are ***,3D finite element models of the proposed composite plates with different layer sequences and thicknesses impacted by a 12.7 mm armor-piercing incendiary(API)are built using Abaqus to predict *** simulation results show that the CFRP/Al2O3 ceramic/Ultrahigh Molecular Weight Polyethylene(UHMWPE)/CFRP(1 mm/4 mm/4 mm/1 mm)composite is the optimized stack of *** simulation results under specified layer sequence and thickness have a reasonable correlation with the experimental results and reflect the failure and fracture of the multi-layer composite protective armor.

关键词： Protective armor composite bionic design structure design mechanical property

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

An origami shield with supporting frame structures optimized by a feature-driven topology optimization method

引用

Defence Technology（防务技术） 2024年第1期31卷 447-456页

作者： Dongsheng Jia Pengcheng Feng Liangdi Wang Longcan Chen Jun Wang jihong zhu Yingjie Xu Weihong Zhang State IJR Center of Aerospace Design and Additive Manufacturing Northwestern Polytechnical UniversityXi'an710072China

In this paper,the design,manufacture and testing of an origami protective shield with a supporting frame structure are *** consists of an origami shield surface and a deployable supporting frame structure that needs to be portable and sufficiently ***,for the design of the shield surface,a threestage origami crease pattern is developed to reduce the shield size in the folded *** shield surface consists of several stiff modular panels and layered with flexible *** modular panels are made of a multi-layer composite where a ceramic layer is made of small pieces to improve durability as those small pieces enable restriction of crack ***,the supporting frame structure is designed as a chain-of-bars structure in order to fold into a highly compact state as a bundle of bars and deploy in ***,a feature-driven topology structural optimization method preserving component sequence is developed where the inter-dependence of sub-structures is taken into account.A bar with semi-circular ends is used as a basic design *** positions of the bar’s end points are treated as design variables and the width of the bars is kept ***,a constraint on the total length of the chain of bars is ***,the modular panels made of multi-layer composite and the full-scale prototype of the origami shield are fabricated and tested to verify the bullet-proof performance.

关键词： Origami Deployable structure Structure design Shield Composite materials

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Learning to bag with a simulation‐free reinforcement learning framework for robots

引用

IET Cyber-Systems and Robotics 2024年第2期6卷 52-66页

作者： Francisco Munguia-Galeano jihong zhu Juan David Hernández Ze Ji Cooper Group University of LiverpoolLiverpoolUK School of Physics Engineering and Technology University of YorkYorkUK School of Computer Science and Informatics Cardiff UniversityCardiffUK School of Engineering Cardiff UniversityCardiffUK

Bagging is an essential skill that humans perform in their daily ***,deformable objects,such as bags,are complex for robots to manipulate.A learning-based framework that enables robots to learn bagging is *** novelty of this framework is its ability to learn and perform bagging without relying on *** learning process is accomplished through a reinforcement learning(RL)algorithm introduced and designed to find the best grasping points of the bag based on a set of compact state *** framework utilises a set of primitive actions and represents the task in five *** our experiments,the framework reached 60% and 80% success rates after around 3 h of training in the real world when starting the bagging task from folded and unfolded states,***,the authors test the trained RL model with eight more bags of different sizes to evaluate its generalisability.

关键词： reinforcement learning robot learning robotics

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

基于LSTM–PPO算法的多机空战智能决策及目标分配

引用

工程科学学报 2024年第7期46卷 1179-1186页

作者：丁云龙匡敏驰朱纪洪祝靖宇乔直新疆大学计算机科学与技术学院乌鲁木齐830000 清华大学精密仪器系北京100084

针对传统多机空战中智能决效率低、难以满足复杂空战环境的需求以及目标分配不合理等问题.本文提出一种基于强化学习的多机空战的智能决策及目标分配方法.使用长短期记忆网络(Long short-term memory,LSTM)对状态进行特征提取和态势感知... 详细信息

针对传统多机空战中智能决效率低、难以满足复杂空战环境的需求以及目标分配不合理等问题.本文提出一种基于强化学习的多机空战的智能决策及目标分配方法.使用长短期记忆网络(Long short-term memory,LSTM)对状态进行特征提取和态势感知,将归一化和特征融合后的状态信息训练残差网络和价值网络,智能体通过近端优化策略(Proximal policy optimization,PPO)针对当前态势选择最优动作.以威胁评估指标作为分配依据,计算综合威胁度,优先将威胁值最大的战机作为攻击目标.为了验证算法的有效性,在课题组搭建的数字孪生仿真环境中进行4v4多机空战实验.并在相同的实验环境下与其他强化学习主流算法进行比较.实验结果表明,使用LSTM–PPO算法在多机空战中的胜率明显优于其他主流强化学习算法,验证了算法的有效性.

关键词：多机空战智能决策近端优化策略威胁评估目标分配

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

考虑接头参数化的空间桁架轻量化设计

引用

航空学报 2024年第5期45卷 352-364页

作者：张瑞桐汪雷刘佳佳朱继宏西北工业大学航宇材料结构一体化设计与增材制造装备技术国际联合研究中心西安710072 中国运载火箭技术研究院北京100076 北京宇航系统工程研究所北京100076

提出了一种考虑接头参数化建模的连杆布局优化方法,该建模过程是基于轴线夹角及尺寸完成的。管状接头结构的几何特性由被连接杆件之间的相互位置和尺寸决定,通过将其相对位置转化为接头轴线夹角,将其尺寸转化为接头主管及支管尺寸实现... 详细信息

提出了一种考虑接头参数化建模的连杆布局优化方法,该建模过程是基于轴线夹角及尺寸完成的。管状接头结构的几何特性由被连接杆件之间的相互位置和尺寸决定,通过将其相对位置转化为接头轴线夹角,将其尺寸转化为接头主管及支管尺寸实现了接头有限元模型的参数化建模。接头结构的空间位置可由被连接件的空间位置转化的轴线交点决定,综上可以实现接头有限元模型的嵌入及整体桁架结构的有限元模型建立。利用基于高斯过程的贝叶斯方法对于桁架连杆布局进行优化设计,同时将制造难度纳入考虑并给出具体的设计方案。数值算例表明,优化方法可有效提高优化目标即结构固有频率。

关键词：布局优化接头设计桁架新型航天器贝叶斯方法

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

全身免疫炎症指数与急性肺栓塞危险分层相关性分析

引用

中华急诊医学杂志 2024年第9期33卷 1286-1290页

作者：杜昌马勇石苗朱继红北京大学人民医院急诊科北京100044

目的分析全身免疫炎症指数(systemic immune inflammation index,SII)与急性肺栓塞危险分层的关系。方法回顾性分析2018年1月至2022年5月北京大学人民医院急诊科抢救室收治的中危急性肺栓塞患者的临床资料,根据危险分层进一步分为中高... 详细信息

目的分析全身免疫炎症指数(systemic immune inflammation index,SII)与急性肺栓塞危险分层的关系。方法回顾性分析2018年1月至2022年5月北京大学人民医院急诊科抢救室收治的中危急性肺栓塞患者的临床资料,根据危险分层进一步分为中高危组急性肺栓塞和中低危组急性肺栓塞,对比两组患者的SII、简化肺栓塞严重度指数(simplified pulmonary embolism severity index,sPESI)以及一般资料、既往史、实验室检查结果等指标,采用Logistic回归分析SII与急性肺栓塞症患者危险分层的关系。结果纳入97例中危急性肺栓塞患者,其中43例为中高危组,54例为中低危组。中高危组患者在SII、白细胞计数、C反应蛋白、sPESI高于中低危组患者(均P<0.05),淋巴细胞计数低于中低危组患者(P<0.05)。Logistic回归分析显示,SII(OR=1.001,95%CI:1.000~1.002,P=0.027)和sPESI(OR=2.000,95%CI:1.155~3.464,P=0.013)是影响急性肺栓塞症患者危险分层的主要因素。ROC曲线分析显示SII、sPESI评分以及两者联合评估急性肺栓塞症危险分层的AUC分别为0.710、0.625、0.739,两者联合评估价值高于单一指标(P=0.007)。结论SII对于急性肺栓塞症危险分层中有良好的应用价值,联合sPESI评分价值更高。

关键词：急性肺栓塞危险分层全身免疫炎症指数简化肺栓塞严重指数

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

基于混合动作的空战分层强化学习决策算法

引用

航空学报 2024年第17期45卷 156-173页

作者：李佐龙朱纪洪匡敏驰张杰任洁清华大学精密仪器系北京100084 航空工业成都飞机设计研究所成都610091

智能空战是世界主要军事强国的研究热点。为解决超视距空战博弈机动决策问题,提出了基于深度强化学习的超视距空战分层决策算法。在该决策算法中,使用适合于超视距空战的机动动作集,对飞机的航迹和姿态进行控制。为了扩大模型的动作空间... 详细信息

智能空战是世界主要军事强国的研究热点。为解决超视距空战博弈机动决策问题,提出了基于深度强化学习的超视距空战分层决策算法。在该决策算法中,使用适合于超视距空战的机动动作集,对飞机的航迹和姿态进行控制。为了扩大模型的动作空间,提升模型的决策能力,将空战的动作空间进行分层,建模为多维离散的动作空间。针对空战中稀疏奖励的问题,设计了一套综合考虑位置优势、武器发射和武器威胁等要素的奖励函数,用于引导智能体向最优策略收敛。搭建了完整的数字孪生空战仿真环境和空战专家系统,在仿真环境中训练决策算法,并通过与专家系统的对抗,对决策算法进行评估。实验结果表明:决策算法具备超视距空战自主决策的能力,能够根据战场态势,进行灵活的机动决策,在与专家系统对抗的过程中具有一定的优势。

关键词：超视距空战智能决策深度强化学习近端策略优化机动动作分层决策

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：