Robotics and Computer-integrated Manufacturing最新文献

{"title":"Dual-service combination optimization of manufacturing and logistics: models for self-managed and third-party logistics in cloud manufacturing","authors":"Chunhua Tang ,&nbsp;Shuangyao Zhao ,&nbsp;Ting Huang ,&nbsp;Mark Goh","doi":"10.1016/j.rcim.2025.103178","DOIUrl":"10.1016/j.rcim.2025.103178","url":null,"abstract":"<div><div>Service combination (SC) is a critical technique in cloud manufacturing, enabling the integration of multiple services to deliver value-added solutions. Logistics plays a pivotal role in SC by ensuring seamless coordination across various manufacturing stages, thereby maximizing the efficiency of production flows. This implies that the SC process must integrate both manufacturing services (MSs) and logistics services (LSs) to determine the optimal combination strategy. Prior research has focused mainly on MS performance, often overlooking the critical impact of logistics on SC outcomes. Although some studies have incorporated logistics considerations, they have largely treated logistics attributes as secondary components of MS evaluations or adopted linear aggregation methods to jointly configure MSs and LSs. These approaches fail to capture the dynamic nature of logistics performance and the interdependencies between MSs and LSs. To address these gaps, this study develops two optimization models for SC that integrate both MSs and LSs, tailored for self-managed and third-party logistics modes. In particular, an innovative bi-level optimization model is introduced to capture the sequential dependencies and dynamic interactions between MSs and LSs in logistics outsourcing, ensuring seamless integration. The upper level focuses on optimizing the MS selection, while the lower level identifies the optimal LSs based on the determined MSs. Improved genetic algorithms incorporating adaptive and parallel mechanisms are developed to address the models, dynamically adjusting parameters to improve solution accuracy and efficiency. Case studies and numerical experiments validate the effectiveness of the proposed models and algorithms, offering actionable managerial insights grounded in the results.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"99 ","pages":"Article 103178"},"PeriodicalIF":11.4,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145614037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From drawings to decisions: A hybrid vision-language framework for parsing 2D engineering drawings into structured manufacturing knowledge","authors":"Muhammad Tayyab Khan ,&nbsp;Lequn Chen ,&nbsp;Zane Yong ,&nbsp;Jun Ming Tan ,&nbsp;Wenhe Feng ,&nbsp;Seung Ki Moon","doi":"10.1016/j.rcim.2025.103186","DOIUrl":"10.1016/j.rcim.2025.103186","url":null,"abstract":"<div><div>Efficient and accurate extraction of key information from 2D engineering drawings is essential for advancing digital manufacturing workflows. This information includes elements such as geometric dimensioning and tolerancing (GD&amp;T), measures, material specifications, and textual annotations. Manual extraction remains slow and labor-intensive, while generic optical character recognition (OCR) models often fail to interpret 2D drawings accurately due to complex layouts, engineering symbols, and rotated annotations. These limitations result in incomplete and unreliable outputs. To address these challenges, this paper proposes a hybrid vision-language framework that integrates a rotation-aware object detection model (YOLOv11-obb) with a transformer-based vision-language parser. We introduce a structured parsing pipeline that first applies YOLOv11-obb to localize annotations and extract oriented bounding box (OBB) image patches, which are subsequently parsed into structured outputs using a fine-tuned, lightweight vision-language model (VLM). To develop and evaluate this pipeline, we curate a dataset of 1367 2D mechanical drawings manually annotated across nine key categories: GD&amp;Ts, General Tolerances, Measures, Materials, Notes, Radii, Surface Roughness, Threads, and Title Blocks. YOLOv11-obb is trained on this dataset to detect OBBs and extract annotation patches. These image patches are then parsed using two fine-tuned open-source VLMs. The first is Donut, a transformer-based model that combines a Swin-B visual encoder with a BART text decoder, enabling end-to-end parsing directly from images without relying on OCR. The second is Florence-2, a prompt-driven encoder–decoder model that integrates a DaViT vision backbone and supports structured output generation through multimodal token alignment. Both models are lightweight and well-suited for specialized industrial tasks under limited computational overhead. Following fine-tuning of both models on the curated dataset of image patches paired with structured annotation labels, a comparative experiment is conducted to evaluate parsing performance across four key metrics. Donut outperforms Florence-2, achieving 89.2 % precision, 99.2 % recall, and a 94 % F1-score, with a hallucination rate of 10.8 %. Finally, a case study demonstrates how the extracted structured information supports downstream manufacturing tasks such as process and tool selection, showcasing the practical utility of the proposed framework in modernizing 2D drawing interpretation.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"99 ","pages":"Article 103186"},"PeriodicalIF":11.4,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145614041","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Pivot-Move Strategy for Dual-Robot Manipulator Additive Manufacturing: Enabling Collision Avoidance without Halting Deposition","authors":"C.L. Li ,&nbsp;Y.C. Jiao ,&nbsp;K. Ren ,&nbsp;N. Liu ,&nbsp;Y.F. Zhang","doi":"10.1016/j.rcim.2025.103177","DOIUrl":"10.1016/j.rcim.2025.103177","url":null,"abstract":"<div><div>Robot-assisted additive manufacturing (AM) has been gaining increasing popularity due to its great flexibility and reachability. Moreover, an AM system with dual deposition-heads held by robot manipulators would significantly shorten the building time, especially for large-scale parts. However, motion planning (MP) for the dual robot manipulators AM is highly challengeable due to various constraints imposed by the setup and the AM process aiming to improve the qualities of the component, e.g., maintaining travelling speed and posture of the deposition head and avoiding collision. In this paper, a novel pivot-move strategy is proposed for MP in AM with dual robot manipulators. Given the sequenced deposition toolpath segments to each deposition head, an initial MP solution including robot configuration at each time sample waypoint is firstly generated for each robot manipulator, respectively. This is followed by conducting a check-and-correct process at each waypoint, where the collision among the links of two robot manipulators is identified and corrected. Specially, the robot manipulator is designed to simultaneously pivot and move to avoid the collision while maintaining the traveling speed unchanged. Numerical simulation, physical implementation, and benchmarking were conducted to exhibit a 78.295% deposition time reduction and high-quality deposition with the developed strategy. To the best of the authors' knowledge, this study represents the pioneering effort in addressing the collision issue in dual robot manipulators depositing on the same heated bed, achieving collision avoidance without interrupting the ongoing deposition process. It can be a valuable supplement to the state of the art in this area.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"99 ","pages":"Article 103177"},"PeriodicalIF":11.4,"publicationDate":"2025-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145611798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A robotic framework for high-throughput and multi-view 3D digital image correlation (3D-DIC): Increasing measurement volume and versatility for deformation analysis","authors":"Özgüç Bertuğ Çapunaman ,&nbsp;Alale Mohseni ,&nbsp;Dennis Dombrovskij ,&nbsp;Kaiyang Yin ,&nbsp;Benay Gürsoy ,&nbsp;Max David Mylo","doi":"10.1016/j.rcim.2025.103187","DOIUrl":"10.1016/j.rcim.2025.103187","url":null,"abstract":"<div><div>Three-dimensional digital image correlation (3D-DIC) is a widely applicable, non-contact optical imaging technique for accurately quantifying full-field surface displacements and strains in materials and structures. However, conventional 3D-DIC implementations relying on fixed stereo camera positions face trade-offs between the field-of-view and spatial resolution and lack high-throughput for long-duration measurements. Here we present an integrated robotic 3D-DIC framework that employs an industrial robotic arm to autonomously and repeatedly reposition stereo cameras. This enables automated calibration, monitoring of multiple samples over extended periods, and expansion of the effective spatial coverage and data throughput, all while maintaining calibration stability and measurement fidelity. We validate this approach on rigid and deforming reference samples and demonstrate its ability to quantify material deformation of bio-composite samples simultaneously during the drying process. Under robotic repositioning, rigid samples exhibit stable displacement and strain measurements while benefiting from significantly increased volumetric coverage and reduced manual oversight. Thus, the proposed system improves experimental efficiency and allows for the incorporation of advanced techniques, such as multi-view stitching, to characterize complex geometries with higher effective resolution. When applied to slowly deforming bio-composites, the system can capture time-lapse images from multiple viewpoints, providing a more comprehensive assessment of complex, evolving material behaviors. These enhancements in 3D-DIC further improve geometric accuracy, increase data density, and expand applicability to a broader range of materials and experimental conditions. Ultimately, the proposed robot-assisted 3D-DIC system creates a robust, high-throughput monitoring framework for bio-fabrication, additive manufacturing, and advanced composite processing, paving the way for targeted programming of shape changes, among other applications.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"99 ","pages":"Article 103187"},"PeriodicalIF":11.4,"publicationDate":"2025-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145595008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A lightweight object detection approach for precision gripping in multiple peg-in-hole assembly tasks","authors":"Jianjun Jiao ,&nbsp;Zonggang Li ,&nbsp;Guangqing Xia ,&nbsp;Guoping Wang ,&nbsp;Yinjuan Chen ,&nbsp;Ruibing Gao","doi":"10.1016/j.rcim.2025.103185","DOIUrl":"10.1016/j.rcim.2025.103185","url":null,"abstract":"<div><div>Automating product assembly using manipulators in manufacturing remains challenging. This is mainly because detection and gripping prior to component assembly still depend heavily on manual operations and traditional teaching methods, resulting in a low overall level of automation. The primary difficulty in detection and gripping arises from the precise recognition of rotation angles and the complex demands for accuracy, real-time performance, and stability. This paper presents an improved lightweight model, IDPC-YOLOv8, for multiple peg-in-hole workpiece detection and gripping to address these challenges. The proposed approach integrates adaptive image preprocessing to enhance visual clarity under varying lighting conditions and employs an efficient network architecture that jointly exploits global and local features to improve detection precision and computational efficiency. In addition, a rotation-aware detection strategy is introduced to enable accurate prediction of object orientation. Moreover, a network optimization scheme further reduces model parameters, making the system suitable for real-time deployment. Experimental results reveal that the IDPC-YOLOv8 model achieves an accuracy of 97.8% and a detection speed of 126.59 FPS, representing improvements of 4% and 8.3%, respectively, over the original YOLOv8-OBB model. Compared to several state-of-the-art rotation detection models, IDPC-YOLOv8 demonstrates superior integration and generalization capabilities. The effectiveness of the proposed method is further validated through excellent gripping success rates achieved in real-world experiments using the AUBO-i5 manipulator.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"98 ","pages":"Article 103185"},"PeriodicalIF":11.4,"publicationDate":"2025-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145567450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robotic disassembly of snap-fit plug connectors in end-of-life electric vehicle batteries","authors":"Jun Huang ,&nbsp;Quanyong Huang ,&nbsp;Yuqin Zeng ,&nbsp;Muyao Tan ,&nbsp;Zhenfeng Peng ,&nbsp;Huawei Song ,&nbsp;Xiuyi Ao ,&nbsp;Duc Pham","doi":"10.1016/j.rcim.2025.103183","DOIUrl":"10.1016/j.rcim.2025.103183","url":null,"abstract":"<div><div>Disassembly is the first step in the remanufacturing of End-of-Life (EoL) Electric Vehicle (EV) batteries. Currently, many disassembly procedures for EV batteries are performed by human operators. Robotic disassembly of EV batteries is essential for increasing the efficiency of the process. A common operation in EV battery disassembly is removing plug connectors. This paper introduces a new method for automating the disassembly of a snap-fit plug connector using a specially designed tool. A control strategy combining force and position was implemented. Experiments were performed on a single connector to investigate and validate the proposed method for disassembling snap-fit plug connectors. The results show that the success rate and integrity rate of the method were both 100 % across 100 tests. Finally, the paper presents a case study on disassembling snap-fit plug connectors in an EV battery. The case study shows that the disassembly approach is feasible and practical, and it can facilitate the automated disassembly of EV batteries.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"98 ","pages":"Article 103183"},"PeriodicalIF":11.4,"publicationDate":"2025-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145567798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel paradigm of robotic machining towards embodied intelligent manufacturing: Case study on paint defect repair","authors":"Shengzhe Wang ,&nbsp;Ziyao Tan ,&nbsp;Yidan Wang ,&nbsp;Zhilei Zhou ,&nbsp;Dahu Zhu","doi":"10.1016/j.rcim.2025.103184","DOIUrl":"10.1016/j.rcim.2025.103184","url":null,"abstract":"<div><div>Embodied intelligence is driving robotic systems towards autonomous capabilities, and its current applications primarily focus on basic tasks such as grasping and navigation. In industrial manufacturing scenarios with complex processes and stringent standards, such as robotic machining, the existing research and applications are less explored. To fill the gap, this paper attempts to propose a novel paradigm of robotic machining from the perspective of embodied intelligence, particularly presenting a robotic grinding process framework to repair paint defects. The framework begins with defect detection using the YOLO algorithm, combined a monocular vision target mapping method to achieve high-precision defect perception. Building upon this, a large language model (LLM), fine-tuned on a process database constructed from empirical defect repair experiments, performs autonomous decision-making for the repair process based on the perceived information. A predefined code library compatible with industrial robot is then developed, enabling the system to automatically generate executable instructions for repair tasks. Both the effectiveness and practicality of the proposed method are validated through a case study on paint defect repair for high-speed train (HST) body.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"98 ","pages":"Article 103184"},"PeriodicalIF":11.4,"publicationDate":"2025-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145560043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-agent deep reinforcement learning for low-carbon flexible job shop scheduling with variable sublots","authors":"Chuanzhao Yu,&nbsp;Youshan Liu,&nbsp;Chunjiang Zhang,&nbsp;Weiming Shen","doi":"10.1016/j.rcim.2025.103180","DOIUrl":"10.1016/j.rcim.2025.103180","url":null,"abstract":"<div><div>As manufacturing shifts toward greener and more intelligent paradigms, traditional scheduling approaches are increasingly inadequate for meeting both operational efficiency and sustainability demands. The Low-Carbon Flexible Job Shop Scheduling Problem with Variable Sublots (LC-FJSP-VS) introduces significant complexity due to the need to simultaneously coordinate sublot sizing, machine selection, and carbon-aware objectives under dynamic disturbances. To address these challenges, this paper proposes a hybrid scheduling framework that integrates Multi-Agent Deep Reinforcement Learning (MADRL) with a bi-objective Mixed-Integer Linear Programming (MILP) model. A hierarchical decision-making architecture is designed, where in the operation-level agent performs real-time job dispatching, and the machine-level agent adjusts processing speeds and optimization preferences to guide sublot-level MILP scheduling. Machine failure events are stochastically simulated to emulate realistic disruptions, testing the system’s adaptability and robustness. Experimental results on extended benchmark datasets show that the proposed method significantly outperforms classical dispatching rules and advanced metaheuristics in terms of Hypervolume (HV), effectively balancing makespan and carbon emissions. This work demonstrates the feasibility and advantages of intelligent, low-carbon scheduling systems and provides a foundation for scalable and disturbance-resilient production planning.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"98 ","pages":"Article 103180"},"PeriodicalIF":11.4,"publicationDate":"2025-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145515809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Intelligent tool wear monitoring approach in milling of titanium alloys","authors":"Shucai Yang,&nbsp;Runjie Jiang,&nbsp;Zekun Song,&nbsp;Dongqi Yu","doi":"10.1016/j.rcim.2025.103181","DOIUrl":"10.1016/j.rcim.2025.103181","url":null,"abstract":"<div><div>Tool wear exerts a critical influence on machining stability and workpiece quality, making its accurate, intelligent monitoring indispensable for preventing tool failure and ensuring product consistency. Although direct assessment via wear imagery is possible, it requires interrupting the machining process and thus is impractical for real‐time production. A more viable solution is to leverage in‐process signals—such as vibration—to enable continuous monitoring. Here, we present a Signal processing method that Beluga whale optimization‐Successive variational mode decomposition (BWO‐SVMD) for noise suppression, followed by the S‐transform to produce high‐resolution time–frequency representations. Based on these denoised spectrograms, we develop an intelligent monitoring model that integrates a multi‐scale convolutional neural network (MSCNN), long short‐term memory (LSTM) units, and a channel–spatial attention mechanism. Experimental results demonstrate that our model achieves 96.25 % classification accuracy, a Kappa coefficient of 0.9686, and a total computation time of 320.64 s. Compared with CNN‐LSTM‐Attention, MSCNN‐Attention, and MSCNN‐LSTM baselines, it improves average accuracy by 1.89 %, 8.02 %, and 6.67 % and Kappa by 0.0732, 0.1374, and 0.2009, respectively. Although training time increases by 10.2 %–14.2 %, the substantial gains in predictive performance justify the additional computational cost.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"98 ","pages":"Article 103181"},"PeriodicalIF":11.4,"publicationDate":"2025-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145498721","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Topology matrix-based interlocking path planning method for robotic additive manufacturing of thin-walled multi-rib structures","authors":"Tao Zhao ,&nbsp;Zhaoyang Yan ,&nbsp;Xiaoyong Zhang ,&nbsp;Runsheng Li ,&nbsp;Kehong Wang ,&nbsp;Shujun Chen","doi":"10.1016/j.rcim.2025.103182","DOIUrl":"10.1016/j.rcim.2025.103182","url":null,"abstract":"<div><div>Thin-walled multi-rib structures are widely used in high-end manufacturing sectors such as aerospace and defense equipment due to their high strength-to-weight ratio. However, traditional manufacturing methods face challenges including prolonged processing cycles and low material utilization. Arc-based directed energy deposition (DED-Arc) technology, characterized by its high efficiency and flexibility, offers a novel approach for the rapid fabrication of thin-walled multi-rib structures. This study focuses on high-ribbed panels in thin-walled multi-rib structures, analyzing their common structural characteristics and proposing a unified path planning method based on an interlocking topology matrix. A standardized topological matrix data structure was developed to describe the medial-axis nodes and topological relationships of high-ribbed panels. A unified path search algorithm was designed based on the topological matrix, employing an alternating search strategy (X-direction for odd layers and Y-direction for even layers) to generate continuous deposition paths. By strategically offsetting the printing contours between adjacent layers, the method achieves topological dispersion and mutual interlocking of weak points across sliced layers. The cross-regions were specifically optimized to ensure overlap-free deposition paths and rational distribution of arc ignition/extinction positions, effectively reducing the number of arc ignition/extinction and improving forming quality. Deposition experiments on four typical thin-walled multi-rib structures demonstrated that the interlocking path planning method significantly enhances surface quality by mitigating height differences at arc ignition/extinction points and improving overlap at intersections., while maintaining overall height errors within 3 mm. The results demonstrate that the proposed method improves manufacturing efficiency and forming quality, supporting DED-Arc applications in lightweight structures.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"98 ","pages":"Article 103182"},"PeriodicalIF":11.4,"publicationDate":"2025-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145509598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}