检索结果-内蒙古大学图书馆

Bin Packing Optimization via Deep Reinforcement Learning

IEEE robotICS AND AUTOMATION LETTERS 2025年第3期10卷 2542-2549页

作者： Wang, Baoying Lin, Zhaohui Kong, Weijie Dong, Huixu Zhejiang Univ Mech Engn Dept Grasp Lab Hangzhou 310027 Peoples R China Zhejiang Key Lab Ind Big Data & Robot Intelligent Hangzhou 310058 Peoples R China

The bin packing problem (BPP) has attracted enthusiastic research interest recently, owing to its widespread applications in logistics and warehousing environments. It is truly essential to optimize the bin packing to enable more objects to be packed into bins, in which the object packing order and placement position are the two crucial optimization goals. However, existing optimization methods for BPP, such as the genetic algorithm (GA), emerge as the primary issues in highly time cost and relatively low accuracy, making it difficult to implement in realistic scenarios. To well relieve related research gaps, we present a novel optimization method of 2D and 3D BPP for objects with regular shapes via deep reinforcement learning (DRL), maximizing the space utilization and minimizing the usage number of bins. First, an end-to-end DRL neural network constructed by a modified Pointer Network consisting of an encoder, a decoder and an attention module is proposed to achieve the optimal object packing order. Second, conforming to the top-down operation mode, the placement strategy based on a height map is used to determine the placement positions of the ordered objects in the bins, preventing the objects from colliding with bins and other objects in bins. Third, the reward and loss functions are defined as the indicators of the compactness, pyramid, and usage number of bins to conduct the DRL neural network training based on an on-policy actor-critic framework. Finally, we conduct extensive experiments to evaluate the performance of the proposed method, and demonstrate that our method achieves a 3% improvement and more than 50x time saving over the GA. Further, an experiment on robotic packing is implemented to validate its generalization capacity in the realistic environment.

关键词： Three-dimensional displays robots Genetic algorithms Costs Deep reinforcement learning Decoding Accuracy Search problems Logistics Convolution Reinforcement learning manipulation planning bin packing problem (BPP) robot packing

来源：评论

学校读者我要写书评

暂无评论

Collaborative graph neural networks for augmented graphs: A local-to-global perspective

引用

PATTERN RECOGNITION 2025年 158卷

作者： Guo, Qihang Yang, Xibei Li, Ming Qian, Yuhua Jiangsu Univ Sci & Technol Sch Econ & Management Zhenjiang 212100 Jiangsu Peoples R China Jiangsu Univ Sci & Technol Sch Comp Zhenjiang 212100 Jiangsu Peoples R China Zhejiang Inst Optoelect Jinhua 321004 Zhejiang Peoples R China Zhejiang Normal Univ Zhejiang Key Lab Intelligent Educ Technol & Applic Jinhua 321004 Zhejiang Peoples R China Shanxi Univ Inst Big Data Sci & Ind Taiyuan 030006 Shanxi Peoples R China

In the field of graph neural networks (GNNs) for representation learning, a noteworthy highlight is the potential of embedding fusion architectures for augmented graphs. However, prevalent GNN embedding fusion architectures mainly focus on handling graph combinations from a global perspective, often ignoring their collaboration with the information of local graph combinations. This inherent limitation constrains the ability of the constructed models to handle multiple input graphs, particularly when dealing with noisy input graphs collected from error-prone sources or those resulting from deficiencies in graph augmentation methods. In this paper, we propose an effective and robust embedding fusion architecture from a local-to-global perspective termed collaborative graph neural networks for augmented graphs (LoGo-GNN). LoGo-GNN). Essentially, LoGo-GNN leverages a pairwise graph combination scheme to generate local perspective inputs. Together with the global graph combination, this serves as the basis to generate a local-to-global perspective. Specifically, LoGo-GNN employs a perturbation augmentation strategy to generate multiple augmentation graphs, thereby facilitating collaboration and embedding fusion from a local-to-global perspective through the use of graph combinations. In addition, LoGo-GNN incorporates a novel loss function for learning complementary information between different perspectives. We also conduct theoretical analysis to assess its expressive power under ideal conditions, demonstrating the effectiveness of LoGo-GNN. Our experiments, focusing on node classification and clustering tasks, highlight the superior performance of LoGo-GNN compared to state-of-the-art methods. Additionally, robustness analysis further confirms its effectiveness in addressing uncertainty challenges.

关键词： Graph neural networks Multi-perspective learning Embedding fusion Complementary learning

来源：评论

学校读者我要写书评

暂无评论

A Path Planning Method Based on Optimal Arm Angle of 7-Degrees-of-Freedom Manipulator

引用

JOURNAL OF intelligent & robotIC SYSTEMS 2025年第2期111卷 1-20页

作者： Zhang, Li Han, Bing Niu, Jianwei Tong, Kun Tian, Xinyang Yu, Xiaolong Hou, Renluan Guo, Yuliang Beihang Univ Hangzhou Innovat Inst Zhejiang Ind Big Data & Robot Intelligent Syst Key Hangzhou 310052 Peoples R China Zhejiang Univ Sch Mech Engn Hangzhou 310030 Peoples R China Beihang Univ Sch Comp Sci & Engn State Key Lab Virtual Real Technol & Syst Beijing 100191 Peoples R China Beihang Univ Sch Automat Sci & Elect Engn Beijing 100191 Peoples R China

This paper aims to address the issue of explicitly utilizing the joint limit avoidance, singularity avoidance, and self-motion obstacle avoidance properties of a 7-degrees-of-freedom manipulator during path planning, given a cartesian space task path. First, the redundant degree of freedom of the manipulator is represented using arm angle parameterization, and the mapping of joint limits and singular configurations to arm angle space is derived in detail, in conjunction with analytical inverse kinematics solution. Next, based on the cartesian space task path, the arm angle space is discretized to obtain the feasible arm angle range under the self-motion obstacle avoidance condition. The arm angle values determine the joint angles of the manipulator. To minimize task execution time and improve efficiency, the total joint angle displacement of the manipulator is set as the optimization objective, and an improved particle swarm optimization algorithm is employed to iteratively search for the optimal arm angles at each path point. Finally, the effectiveness of the proposed method is validated through both simulation and real machine experiments.

关键词： Redundant manipulator Inverse kinematics Path planning Joint limit avoidance Self-motion obstacle avoidance

来源：评论

学校读者我要写书评

暂无评论

Towards reliable and damage-less robotic fragile fruit grasping: An enveloping gripper with multimodal strategy inspired by Asian elephant trunk

引用

COMPUTERS AND ELECTRONICS IN AGRICULTURE 2025年 234卷

作者： Wang, Qingyu Bai, Kaixin Zhang, Lei Sun, Zhizhong Jia, Tianze Hu, Dong Li, Qiang Zhang, Jianwei Knoll, Alois Jiang, Huanyu Zhou, Mingchuan Ying, Yibin Zhejiang Univ Coll Biosyst Engn & Food Sci Hangzhou 310058 Zhejiang Peoples R China Zhejiang Univ Zhejiang Key Lab Intelligent Sensing & Robot Agr Hangzhou 310058 Peoples R China Minist Agr & Rural Affairs Key Lab Site Proc Equipment Agr Prod Beijing 310058 Peoples R China Univ Hamburg Dept Informat TAMS Tech Aspects Multimodal Syst D-22527 Hamburg Germany Agile Robots D-81477 Munich Germany Zhejiang A&F Univ Coll Chem & Mat Engn Hangzhou 311300 Zhejiang Peoples R China Zhejiang A&F Univ Coll Opt Mech & Elect Engn Hangzhou 311300 Zhejiang Peoples R China Shenzhen Technol Univ Coll Big Data & Internet Shenzhen 518118 Guangdong Peoples R China Tech Univ Munich Dept Comp Sci D-999035 Munich Germany

Fruit uploading and packaging are labor-intensive and time-consuming steps in postharvest industry, which involve continuous pick-and-place manipulation. In this case, we aim to replace manual working with robotic grasping. However, for robotic fragile fruit grasping, the main difficulty is to reduce early stage bruise while maintaining the grasping reliability. In this study, we aim to solve this problem and achieve reliable and damageless robotic fragile fruit grasping. Inspired by the structure of the Asian elephant trunk with its feeding behavior, a bionic and pneumatic soft gripper was designed, and a multimodal grasping strategy was proposed. Similar with Asian elephant trunk, the gripper has two designed trapezoid air chambers to control the two individual parts, including fingertip-like process and enveloping structure. Enveloping grasping behavior was imitated with larger area of contact, less contact force, and larger pull off force. A visuo-tactile multimodal grasping strategy was integrated into the robotic grasping system. The visual modality was developed for positioning and grasp pose estimation. The tactile modality was employed for grasping pose confirmation and closed-loop grasping force control. In the experiment on the enveloping gripper, the maximum contact force and the pull off force reached a good balance and were 0.7083 N and 7.959 N, respectively. With the proposed multimodal grasping strategy, the grasping success rate increased 4.23 % to 96.70 %. As for closed-loop control of the grasping force, the average value for steady-state error and maximum overshoot were 0.0856 N and 26.43 %, respectively. The experiment on Spatial Frequency Domain Imaging (SFDI) demonstrated the effectiveness of our enveloping gripper in reducing the early stage bruise. To some extent, the designed enveloping gripper with the proposed multimodal strategy could achieve reliable and damage-less fragile fruit grasping, which is promising in fruit postharvest industr

关键词： robotic grasping Bionic design Visuo-tactile perception Pneumatic-driven soft gripper Postharvest industry

来源：评论

学校读者我要写书评

暂无评论

An asynchronous federated learning-assisted data sharing method for medical blockchain

引用

APPLIED INTELLIGENCE 2025年第2期55卷 1-21页

作者： Gan, Chenquan Xiao, Xinghai Zhang, Yiye Zhu, Qingyi Bi, Jichao Jain, Deepak Kumar Saini, Akanksha Chongqing Univ Posts & Telecommun Sch Cyber Secur & Informat Law Chongqing 400065 Peoples R China Chongqing Univ Canc Hosp Chongqing 400030 Peoples R China Zhejiang Inst Ind & Informat Technol Hangzhou 310000 Peoples R China ChongQing Univ Sch Big Data & Software Engn Chongqing 400044 Peoples R China Dalian Univ Technol Key Lab Intelligent Control & Optimizat Ind Equipm Minist Educ Dalian 116024 Peoples R China Symbiosis Int Univ Symbiosis Inst Technol Pune 412115 India RMIT Univ Coll Business & Law Melbourne Vic 3000 Australia

Currently, medical blockchain data sharing methods that rely on federated learning face challenges, including node disconnection, vulnerability to poisoning attacks, and insufficient consideration of conflicts of interest among participants. To address these issues, we propose a novel method for data sharing in medical blockchain systems based on asynchronous federated learning. First, we develop an aggregation algorithm designed specifically for asynchronous federated learning to tackle the problem of node disconnection. Next, we introduce a Proof of Reputation (PoR) consensus algorithm and establish a consensus committee to mitigate the risk of poisoning attacks. Furthermore, we integrate a tripartite evolutionary game model to examine conflicts of interest among publishing nodes, committee nodes, and participating nodes. This framework enables all parties involved to make strategic decisions that promote sustainable data-sharing practices. Finally, we conduct a security analysis to validate the theoretical effectiveness of the proposed method. Experimental evaluations using real medical datasets demonstrate that our method outperforms existing approaches.

关键词： Medical blockchain data sharing Asynchronous federated learning Proof of reputation Tripartite evolutionary game

来源：评论

学校读者我要写书评

暂无评论

LUD-YOLO: A novel lightweight object detection network for unmanned aerial vehicle

引用

INFORMATION SCIENCES 2025年 686卷

作者： Fan, Qingsong Li, Yiting Deveci, Muhammet Zhong, Kaiyang Kadry, Seifedine Huazhong Univ Sci & Technol State Key Lab Intelligent Mfg Equipment & Technol Wuhan 430074 Peoples R China Guizhou Univ Finance & Econ Coll Big Data Stat Guiyang 550025 Peoples R China Guizhou Univ Key Lab Adv Mfg Technol Minist Educ Guiyang 550025 Guizhou Peoples R China Natl Def Univ Turkish Naval Acad Dept Ind Engn TR-34942 Istanbul Turkiye Imperial Coll London Royal Sch Mines South Kensington Campus London SW7 2AZ England Western Caspian Univ Dept Informat Technol Baku 1001 Azerbaijan Zhejiang Univ Coll Informat Sci & Elect Engn Hangzhou 310058 Peoples R China Lebanese Amer Univ Dept Comp Sci & Math Beirut Lebanon Middle East Univ MEU Res Unit Amman 11831 Jordan

Autonomous execution of tasks by unmanned aerial vehicles (UAVs) relies heavily on object detection. However, object detection in most images presents challenges such as complex backgrounds, small targets, and obstructions. Additionally, the limited computing speed and memory of the UAV processor affects the accuracy of conventional object detection algorithms. This paper proposes LUD-You Only Look Once (YOLO), a small and lightweight object detection algorithm for UAVs based on YOLOv8. The proposed algorithm introduces a new multiscale feature fusion mode that solves the degradation in feature propagation and interaction through the introduction of upsampling in the feature pyramid network and the progressive feature pyramid network. The application of the dynamic sparse attention mechanism in the Cf2 module achieves flexible computing allocation and content awareness. Furthermore, the proposed model is optimized to be sparse and lightweight, making it possible to deploy on UAV edge devices. Finally, the effectiveness and superiority of LUD-YOLO were verified on the VisDrone2019 and UAVDT datasets. The results of ablation and comparison experiments show that compared with the original algorithm, LUDY-N and LUDY-S have shown excellent performance in various evaluation indexes, indicating that the proposed improvement strategies make the model have better robustness and generalization. Moreover, compared with multiple other popular competitors, the proposed improvement strategies enable LUD-YOLO to have the best overall performance, providing an effective solution for UAVs object detection while balancing model size and detection accuracy.

关键词： Small object detection YOLOv8 UAV Deep learning Feature fusion

来源：评论

学校读者我要写书评

暂无评论

Genetic mapping of serum metabolome to chronic diseases among Han Chinese

引用

CELL GENOMICS 2025年第2期5卷 100743-100743页

作者： Cheng, Chunxiao Xu, Fengzhe Pan, Xiong-Fei Wang, Cheng Fan, Jiayao Yang, Yunhaonan Liu, Yuanjiao Sun, Lingyun Liu, Xiaojuan Xu, Yue Zhou, Yuan Xiao, Congmei Gou, Wanglong Miao, Zelei Yuan, Jiaying Shen, Luqi Fu, Yuanqing Sun, Xiaohui Zhu, Yimin Chen, Yuming Pan, An Zhou, Dan Zheng, Ju-Sheng Zhejiang Univ Affiliated Hosp 2 Sch Publ Hlth Sch Med Hangzhou 310058 Peoples R China Key Lab Intelligent Prevent Med Zhejiang Prov Hangzhou 310058 Zhejiang Peoples R China Westlake Univ Ctr Infect Dis Res Sch Med Zhejiang Key Lab Multi Infect & Immun Hangzhou 310024 Peoples R China Westlake Ctr Intelligent Prote Westlake Lab Life Sci & Biomed Hangzhou 310024 Peoples R China Sichuan Univ West China Univ Hosp 2 Sect Epidemiol & Populat Hlth Chengdu 610041 Peoples R China Sichuan Univ West China Univ Hosp 2 Dept Gynecol & Obstet Key Lab Birth Defects & Related Dis Women & Childr Chengdu 610041 Peoples R China Sichuan Univ West China Univ Hosp 2 Key Lab Tech Res Drug Prod Vitro & Vivo Correlat Natl Med Prod Adm Chengdu 610041 Peoples R China Shuangliu Maternal & Child Hlth Hosp Shuangliu Inst Womens & Childrens Hlth Chengdu 610200 Peoples R China Sichuan Univ West China Hosp West China Biomed Big Data Ctr Chengdu 610041 Peoples R China Sun Yat Sen Univ Sun Yat Sen Mem Hosp Dept Clin Nutr Guangzhou Peoples R China Zhejiang Univ Sch Med Dept Epidemiol & Biostat Hangzhou Zhejiang Peoples R China Sichuan Univ West China Univ Hosp 2 Dept Lab Med Minist EducKey Lab Birth Defects & Related Dis Wo Chengdu 610041 Peoples R China Shuangliu Maternal & Child Hlth Hosp Dept Sci & Educ Chengdu 610200 Sichuan Peoples R China Zhejiang Chinese Med Univ Sch Publ Hlth Hangzhou 310053 Peoples R China Sun Yat Sen Univ Sch Publ Hlth Dept Epidemiol Guangdong Prov Key Lab Food Nutr & Hlth Guangzhou Peoples R China Huazhong Univ Sci & Technol Tongji Med Coll Sch Publ Hlth Dept Epidemiol & BiostatMinist EducKey Lab Envir Wuhan 430030 Peoples R China Westlake Inst Adv Study Inst Basic Med Sci Hangzhou 310024 Peoples R China Westlake Univ Res Ctr Ind Future Sch Life Sci Hangzhou 310024 Peoples R China Westlake Univ Affiliated Hangzhou Peoples Hosp 1 Sch Med Hangzhou Peoples R China

Serum metabolites are potential regulators for chronic diseases. To explore the genetic regulation of metabolites and their roles in chronic diseases, we quantified 2,759 serum metabolites and performed genomewide association studies (GWASs) among Han Chinese individuals. We identified 184 study-wide significant (p <1.81 3 10-11) metabolite quantitative trait loci (metaboQTLs), 88.59% (163) of which were novel. Notably, we identified Asian-ancestry-specific metaboQTLs, including the SNP rs2296651 for taurocholic acid and taurochenodesoxycholic acid. Leveraging the GWAS for 37 clinical traits from East Asians, Mendelian randomization analyses identified 906 potential causal relationships between metabolites and clinical traits, including 27 for type 2 diabetes and 38 for coronary artery disease. Integrating genetic regulation of the transcriptome and proteome revealed putative regulators of several metabolites. In summary, we depict a landscape of the genetic architecture of the serum metabolome among Han Chinese and provide insights into the role of serum metabolites in chronic diseases.

关键词： metabolites genome-wide association study East Asians Mendelian randomization

来源：评论

学校读者我要写书评

暂无评论

PS-TTL: Prototype-based Soft-labels and Test-Time Learning for Few-shot Object Detection 24

PS-TTL: Prototype-based Soft-labels and Test-Time Learning f...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Gao, Yingjie Zhang, Yanan Huang, Ziyue Liu, Nanqing Huang, Di SKLCCSE Beihang University Beijing China SIST Southwest Jiaotong University Chengdu China Zhejiang Industrial Big Data and Robot Intelligent System Key Lab. Hangzhou China

ISBN: (纸本)9798400706868

In recent years, Few-Shot Object Detection (FSOD) has gained widespread attention and made significant progress due to its ability to build models with a good generalization power using extremely limited annotated data. The fine-tuning based paradigm is currently dominating this field, where detectors are initially pre-trained on base classes with sufficient samples and then fine-tuned on novel ones with few samples, but the scarcity of labeled samples of novel classes greatly interferes precisely fitting their data distribution, thus hampering the performance. To address this issue, we propose a new framework for FSOD, namely Prototype-based Soft-labels and Test-Time Learning (PS-TTL). Specifically, we design a Test-Time Learning (TTL) module that employs a mean-teacher network for self-training to discover novel instances from test data, allowing detectors to learn better representations and classifiers for novel classes. Furthermore, we notice that even though relatively low-confidence pseudo-labels exhibit classification confusion, they still tend to recall foreground. We thus develop a Prototype-based Soft-labels (PS) strategy through assessing similarities between low-confidence pseudo-labels and category prototypes as soft-labels to unleash their potential, which substantially mitigates the constraints posed by few-shot samples. Extensive experiments on both the VOC and COCO benchmarks show that PS-TTL achieves the state-of-the-art, highlighting its effectiveness. The code and model are available at https://***/gaoyingjay/PS-TTL. © 2024 ACM.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Hot cracks in camshaft casting: initiation and propagation

引用

INTERNATIONAL JOURNAL OF CAST METALS RESEARCH 2024年第1期37卷 48-70页

作者： Zhu, Yankang Xu, Gangqiang Xu, Cangsu Li, Xiaolu Zhou, Kangkang Chen, Yuan Li, Yuntang Xu, Xinsheng Liu, Hui China Jiliang Univ Coll Mech & Elect Engn Key Lab Intelligent Mfg Qual Big Data Tracing & An Hangzhou Peoples R China Zhejiang Boxing Ind & Trade Co Ltd Ctr Enterprise Technol Jinhua Peoples R China Zhejiang Univ Coll Energy Engn State Key Lab Clean Energy Utilizat Hangzhou Peoples R China China Jiliang Univ Coll Qual & Safety Engn Hangzhou Peoples R China China Jiliang Univ Coll Mech & Elect Engn Key Lab Intelligent Mfg Qual Big Data Tracing & An Hangzhou 310018 Peoples R China

In this paper, a numerical simulation method based on the camshaft casting process and crack expansion was proposed to study the propagation characteristics of casting hot cracks. The dynamic visualisation process of hot crack growth indicates that the formation of hot tear is dominated by opening mode cracks. And during the propagation process, the equivalent stress intensity factor at the crack front first decreases and then increases. Furthermore, the propagation characteristics of the crack under different hot tear initiation conditions were studied by this method. The results reveal that the expansion ability of a hot crack is affected by the equilibrium solidification scale, solidification sequence, and solidification path of the casting. Finally, using the microscopic morphology of the cracks, the semi-quantitative analysis of the elements illustrates that the carbon content decreases sharply at the crack formation site, while manganese and sulphur are relatively rich.

关键词： Camshaft casting process hot crack crack initiation crack propagation stress intensity factor casting solidification mode

来源：评论

学校读者我要写书评

暂无评论

A multi-rate high-order dynamic twin-latent-variable probabilistic modeling and its process monitoring application

引用

ISA TRANSACTIONS 2024年 149卷 281-294页

作者： Ying, Ze Chang, Yuqing He, Yuchen Wang, Fuli Northeastern Univ Coll Informat Sci & Engn Shenyang 110819 Peoples R China Northeastern Univ State Key Lab Synthet Automat Proc Ind Shenyang 110819 Peoples R China China Jiliang Univ Key Lab Intelligent Mfg Qual Big Data Tracing & An Hangzhou 310018 Zhejiang Peoples R China

Quality-relevant process monitoring provides important guarantees for the safety of industrial operation, which is based on the assumption that data collection is complete and low-order autocorrelated. However, real industrial processes always exhibit complex characteristics such as multi-rate sampling and high-order dynamic, which pose great challenges for process monitoring. To this end, a multi-rate high-order dynamic twin-latentvariable probabilistic (MHDTVP) model is presented in this paper to extract data correlations among multirate measurements from quality-relevant and irrelevant perspectives. Moreover, to reveal the dynamics in the multi-rate sampling process, an autoregressive twin-latent-variable structure is designed to extract both qualityrelevant and quality-irrelevant high-order dynamic features. In the MHDTVP model, parameters are trained through an efficient expectation maximization (EM) iteration framework. Finally, the performance conclusions of MHDTVP are validated with the Tennessee Eastman process (TEP) and Thermal Power Plant (TPP). The experimental results demonstrate that the proposed model exhibits superior monitoring efficiency for multi-rate dynamic processes compared to similar approaches.

关键词： Quality -relevant process monitoring Multi -rate sampling process High -order dynamics Probability latent variable model Expectation Maximization (EM) algorithm

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：