Advanced Engineering Informatics最新文献

A context-aware KG-LLM collaborated conceptual design approach for personalized products: A case in lower limbs rehabilitation assistive devices 情境感知KG-LLM合作的个性化产品概念设计方法：以下肢康复辅助装置为例

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-03 DOI: 10.1016/j.aei.2025.103422

Xinyu Pan, Weibin Zhuang, Sijie Wen, Weigang Yu, Jinsong Bao, Xinyu Li

{"title":"A context-aware KG-LLM collaborated conceptual design approach for personalized products: A case in lower limbs rehabilitation assistive devices","authors":"Xinyu Pan, Weibin Zhuang, Sijie Wen, Weigang Yu, Jinsong Bao, Xinyu Li","doi":"10.1016/j.aei.2025.103422","DOIUrl":"10.1016/j.aei.2025.103422","url":null,"abstract":"<div><div>With the rapid increase in demand for personalized Rehabilitation Assistive Devices (RADs), significant challenges have emerged in their design processes. Particularly in practical applications, designers face challenges such as ambiguity in user requirements, inefficiencies in cross-domain knowledge sharing, and deviations of generated solutions from actual user needs. To address these issues, this paper proposes a Context-Aware Conceptual design method based on Knowledge graph (KG) and Large language models (LLM), named CACKL. Firstly, to address the high complexity involved in eliciting user requirements, user profiles are constructed by integrating multi-source data, and fine-grained “requirement-function” mappings are extracted using fine-tuned LLM, thereby reducing the cost associated with manual intervention. Secondly, a KG-LLM collaborated reasoning mechanism guided by a Chain-of-Thought (CoT) prompting approach is proposed to align structured domain knowledge with implicit semantic representations from LLM, thus enhancing the contextual relevance and practical effectiveness of concept generation, aiming to improve the efficiency of personalized conceptual design. In a practical case involving lower-limb RADs, the proposed CACKL method was evaluated regarding user requirement mining and conceptual design. Experimental results demonstrated significant advantages in the automatic generation of personalized design solutions, particularly in enhancing design efficiency and meeting user requirements, thereby validating its effectiveness in real-world applications. This study provides an innovative paradigm for the intelligent design of RADs by integrating dynamic knowledge constraints with natural language interaction.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103422"},"PeriodicalIF":8.0,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A framework for automatic Real-Time Pixel-Level segmentation of underwater dam concrete cracks utilizing the CRTransU-Net model 基于CRTransU-Net模型的水下大坝混凝土裂缝实时自动像素级分割框架

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-03 DOI: 10.1016/j.aei.2025.103415

Yunlin Ma , Tengfei Bao , Yangtao Li , Mengfan Zhao

{"title":"A framework for automatic Real-Time Pixel-Level segmentation of underwater dam concrete cracks utilizing the CRTransU-Net model","authors":"Yunlin Ma , Tengfei Bao , Yangtao Li , Mengfan Zhao","doi":"10.1016/j.aei.2025.103415","DOIUrl":"10.1016/j.aei.2025.103415","url":null,"abstract":"<div><div>To address the challenges in detecting concrete cracks in the underwater sections of hydropower station dams, which are prone to interference from water disturbances and light refraction, a real-time pixel-level automatic segmentation framework for underwater dam concrete cracks is proposed. The architecture adopts a symmetric network structure with skip connections across network layers to enhance feature transmission. A combination strategy of ViT and CBAM is employed to extract complex crack effectively features underwater. Lightweight optimization of the network is achieved by integrating channel pruning and Knowledge Distillation techniques. Additionally, Dice Loss is used to optimize the loss function, overcoming the imbalanced foreground and background issues in underwater crack segmentation. The proposed CRTransU-Net model demonstrates the accurate identification of underwater crack regions. Based on an experimental study conducted on an RCC gravity dam project, the method achieved optimal segmentation performance compared to models such as U-Net, U-Net++, FCN, and DeepLabv3+. The model’s mIoU, Recall, Precision, F1-score, PA, and SM values are 0.90127, 0.95867, 0.95449, 0.94676, 0.94218, and 0.92887, respectively. Furthermore, the geometric dimensions of cracks were quantified by combining regional pixel extraction with infrared laser ranging technology. The quantitative results obtained from the predicted masks fit well with those derived from annotated masks.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103415"},"PeriodicalIF":8.0,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899797","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Analyzing mandatory and discretionary lane change interaction patterns using hidden Markov model-based approaches 使用基于隐马尔可夫模型的方法分析强制和自由变道交互模式

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-02 DOI: 10.1016/j.aei.2025.103404

Yajie Zou , Shubo Wu , Lusa Ding , Yue Zhang , Siyang Zhang , Lingtao Wu

{"title":"Analyzing mandatory and discretionary lane change interaction patterns using hidden Markov model-based approaches","authors":"Yajie Zou , Shubo Wu , Lusa Ding , Yue Zhang , Siyang Zhang , Lingtao Wu","doi":"10.1016/j.aei.2025.103404","DOIUrl":"10.1016/j.aei.2025.103404","url":null,"abstract":"<div><div>It is indispensable for autonomous vehicles (AVs) to understand the complex and dynamic lane-changing interaction patterns, which can support AVs in making appropriate driving decisions. This study proposed a learning framework for understanding the interaction patterns during mandatory lane change (MLC) and discretionary lane change (DLC). Three hidden Markov model (HMM) based approaches, namely HMM with Gaussian mixture model (GMM-HMM), hierarchical Dirichlet process-hidden semi-Markov model (HDP-HSMM), and coupled HMM (CHMM) are compared for segmenting driving primitives. Then dynamic time warping distance-based K-means clustering is employed to group the driving primitives into 6 and 8 interaction patterns for MLC and DLC. The minimum time to collision (TTC) of two conflict types between interactive vehicles involved in the lane-changing scenario is applied to evaluate the traffic risk associated with interaction patterns. Two types of lane-changing events are extracted at a freeway entrance ramp from the international, adversarial, and cooperative motion (INTERACTION) dataset. The experimental results demonstrate that the HDP-HSMM achieves better performance in separating driving primitives with interpretable semantic information, enabling a comprehensive understanding of the dynamic spatiotemporal characteristics and the traffic risk evolution mechanisms of lane-changing interaction patterns. Additionally, the traffic risk associated with interaction patterns of DLC is generally higher than that of MLC. The findings of this study are beneficial for AVs in understanding the collision risk during lane changes, thereby enhancing driving decision-making.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103404"},"PeriodicalIF":8.0,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Development of a semantic segmentation network for plane layout design of super high-rise building structures 超高层建筑结构平面布局设计语义分割网络的开发

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-02 DOI: 10.1016/j.aei.2025.103396

Zhonghui Zhao , Zheng He , Dianyou Yu , Shuyu Tian

{"title":"Development of a semantic segmentation network for plane layout design of super high-rise building structures","authors":"Zhonghui Zhao , Zheng He , Dianyou Yu , Shuyu Tian","doi":"10.1016/j.aei.2025.103396","DOIUrl":"10.1016/j.aei.2025.103396","url":null,"abstract":"<div><div>Development of intelligent structural design (ISD) of buildings has gained increasing attention. The need of ISD for complex super high-rise buildings with inherent substantial challenges is particularly more pressing. As the fundamental step of ISD, a properly-designed semantic segmentation network is essential for extracting the pixel-level sparse information contained in the dramatically variable structural plane layouts along height. As the result of the variations, the segmentation networks developed to date fail to achieve this goal both accurately and efficiently. The well-structured DeepLabv3+ is chosen as the baseline network on which some significant modifications are made to develop Tall-DeepLabv3+, i.e. the replacement of the encoding backbone with a multi-scale convolutional attention network and the integration of three external attention modules, three skip connections and a unified loss function. The accuracy, stability and generalization ability of Tall-DeepLabv3+ is systematically demonstrated through the pre-training, transfer training, ablation experiments and comparative validation analysis. Utilizing a parametrically-generated dataset for frame core-tube building structures, the network achieved a peak mean intersection over union of 93.52 % with a minimal deviation of 0.002. Comparative validation analysis results demonstrated the superior overall performance and segmentation type-level accuracy of Tall-DeepLabv3+ in processing plane layouts characterized by prominent sparsity features.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103396"},"PeriodicalIF":8.0,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895198","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exemplar-free class incremental learning for rotating machinery fault diagnosis via adaptive prototype correction and separation network 基于自适应原型校正和分离网络的无样本类增量学习旋转机械故障诊断

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-02 DOI: 10.1016/j.aei.2025.103420

Zongzhen Ye , Jun Wu , Xuesong He , Lixiang Wang , Weixiong Jiang

{"title":"Exemplar-free class incremental learning for rotating machinery fault diagnosis via adaptive prototype correction and separation network","authors":"Zongzhen Ye , Jun Wu , Xuesong He , Lixiang Wang , Weixiong Jiang","doi":"10.1016/j.aei.2025.103420","DOIUrl":"10.1016/j.aei.2025.103420","url":null,"abstract":"<div><div>Class incremental learning technologies have been extensively studied in rotating machinery fault diagnosis to continuously learn and integrate new diagnosis knowledge. However, existing approaches usually require retaining some historical samples to replay previous diagnosis knowledge when learning new fault categories, which not only poses data privacy leakage but also wastes significant storage and training resources. To address these challenges, a novel Adaptive Prototype Correction and Separation Network (APCSN) is developed for exemplar-free incremental fault diagnosis. In the APCSN, an optimal transport theory-based prototype correction module is designed to adaptively transport historical category prototypes to the new representation space, effectively mitigating the prototype drift caused by model evolution. In addition, a hybrid contrastive learning module that incorporates the old category prototypes and new category features is designed to enhance intra-category feature compactness and inter-category feature separability, thus alleviating the feature overlap between new and old categories. Experiments conducted on the bearing dataset and wind turbine gearbox dataset demonstrate that the APCSN attains the diagnosis accuracy of 99.01% and 97.36%, respectively, outperforming state-of-the-art methods. The results indicate that the APCSN exhibits superior performance in incremental fault diagnosis without reserved old samples.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103420"},"PeriodicalIF":8.0,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895197","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Graph-Based approach for individual fall risk assessment through a wearable inertial measurement unit sensor 基于图的可穿戴惯性测量单元传感器个人跌倒风险评估方法

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-02 DOI: 10.1016/j.aei.2025.103413

Hoonyong Lee , John Sohn , Gaang Lee , Jesse V. Jacobs , SangHyun Lee

{"title":"A Graph-Based approach for individual fall risk assessment through a wearable inertial measurement unit sensor","authors":"Hoonyong Lee , John Sohn , Gaang Lee , Jesse V. Jacobs , SangHyun Lee","doi":"10.1016/j.aei.2025.103413","DOIUrl":"10.1016/j.aei.2025.103413","url":null,"abstract":"<div><div>Exposure to slip, trip, and fall (STF) hazards can serve as a precursor of fall incidents in people’s daily lives. Wearable inertial measurement unit (IMU) sensors have been used to monitor an individual’s body movements for assessing fall risks by detecting abnormal body movements. However, the current models have relied on prior knowledge (e.g., predetermined IMU patterns or pre-trained models) and may therefore fail to generalize across untrained individuals, tasks, and STF hazard exposures. To this end, the authors propose a graph-based approach. By transforming time-series IMU data into a graph structure, in which each data point is represented as a node and the relationships between points are represented as edges, the nonlinear and complex relationships among data points can be captured, allowing the accurate detection of abnormal subsequences in the IMU data without relying on labeled training data. In this study, the degree of IMU signal abnormality while walking is interpreted as exposure to an STF hazard. To test the graph-based STF hazard index, 16 young, healthy subjects walked a laboratory course that included STF hazards. The proposed index averaged 0.90 precision to detect STF hazard exposures, and STF hazard index values yielded an average correlation of 0.95 with the subjects’ self-reported fall risk perceptions of the STF hazards. These results demonstrate the feasibility of the proposed approach to assess fall risk without relying on labeled training data. Thus, with further field research, this approach offers thepotential for large-scale implementation in people’s daily lives.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103413"},"PeriodicalIF":8.0,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895196","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Data-model interaction-driven transferable graph learning method for weak-shot onsite FTU health condition assessment 弱弹现场FTU健康状况评估的数据模型交互驱动可转移图学习方法

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-01 DOI: 10.1016/j.aei.2025.103364

Fengyuan Zhang , Jie Liu , Haoliang Li , Ran Duan , Zhongxu Hu , Tielin Shi

{"title":"Data-model interaction-driven transferable graph learning method for weak-shot onsite FTU health condition assessment","authors":"Fengyuan Zhang , Jie Liu , Haoliang Li , Ran Duan , Zhongxu Hu , Tielin Shi","doi":"10.1016/j.aei.2025.103364","DOIUrl":"10.1016/j.aei.2025.103364","url":null,"abstract":"<div><div>The large amount of monitoring data provided by the onsite hydropower unit has promoted the development of data-driven Francis turbine unit (FTU) health condition assessment (HCA) technology. However, these methods are usually trained in fully annotated source domains and applied to sparse onsite scenarios, leading to weak-shot learning problems. To fully explore the potential state representation from the source domain annotated by the mechanism simulation model, an innovative knowledge graph-based data-model-interaction framework is proposed for solving weak-shot onsite FTU health condition assessment. First, based on the selected critical onsite monitoring data, the pseudo-data obtained from the computational fluid dynamics calculation of the mechanism digital-twin (DT) model are used to fully annotate source domain. Secondly, the mixed pseudo-actual data are converted into graphs by node similarities to capture the correlations between signals. The explicit edge connection relationships in the graph structure allow state sharing across domain nodes and suppress loss of accuracy due to differences in domain distribution. Then, a transferable graph constructor with cross-domain parameter sharing is designed to learn knowledge-based construction strategies from the fully annotated source domain. Further, the transfer of state knowledge from theoretical domain to actual domain can further strengthen the sample’s representation in weak-shot domain. Finally, a graph-driven health benchmark model (HBM) is constructed to excavate the reconstruction-enhanced knowledge graphs, achieving FTU state presentation and degradation assessment. The proposed method has been applied in a dataset collected from onsite FTU, which not only achieves the best performance in multiple SOTA comparison tests, but also has an acceptable time consumption (5.24 s/100 graphs), and has the possibility of industrial field deployment.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103364"},"PeriodicalIF":8.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143887536","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-process digital twin closed-loop machining through shape-feature state update and error propagation knowledge graph 基于形状特征状态更新和误差传播知识图谱的多工序数字双闭环加工

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-01 DOI: 10.1016/j.aei.2025.103403

Zhiwen Lin , Yueze Zhang , Chuanhai Chen , Jinyan Guo , Baobao Qi , Jun Yan , Zhifeng Liu

{"title":"Multi-process digital twin closed-loop machining through shape-feature state update and error propagation knowledge graph","authors":"Zhiwen Lin , Yueze Zhang , Chuanhai Chen , Jinyan Guo , Baobao Qi , Jun Yan , Zhifeng Liu","doi":"10.1016/j.aei.2025.103403","DOIUrl":"10.1016/j.aei.2025.103403","url":null,"abstract":"<div><div>Machining digital twin systems represent a novel platform for achieving full autonomy in machine tools. However, in multi-process machining, error propagation between different processes presents significant challenges to current digital twin systems: real-time model updates, error evolution analysis and optimization. To address these issues, this study proposes a method for rapid state updates of shape features and the construction of an error propagation knowledge graph, enabling dynamic control in a multi-process machining digital twin system. First, a state update method based on multi-level voxel computation is introduced, utilizing the transformation relationship between cutter-workpiece engagement (CWE) and voxels to ensure rapid updates of multi-process digital twin models. Second, digital threads for multi-process machining are developed to identify machining states and track errors. To analyze error evolution and implement effective control measures, an error propagation knowledge graph based on Taylor expansion model is constructed, enabling closed-loop control. Finally, the proposed system was validated for its advancement in error control and closed-loop optimization through the machining of typical multi-process components in the transportation field. The results of comparative experiments and ablation experiments indicate that the error updating efficiency based on multi-level voxel computation threads improved by an average of 59.26% compared to other benchmarks. The error propagation knowledge graph based on the Taylor model achieved a 42.75% improvement in reasoning accuracy compared to the ablated Taylor model. The developed digital twin closed-loop optimization system successfully ensured that the errors in 1280 processes of the case study remained within the tolerance range.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103403"},"PeriodicalIF":8.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143894993","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MAPPO-ITD3-IMLFQ algorithm for multi-mobile robot path planning 多移动机器人路径规划的MAPPO-ITD3-IMLFQ算法

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-01 DOI: 10.1016/j.aei.2025.103398

Likun Hu, Chunyou Wei, Linfei Yin

{"title":"MAPPO-ITD3-IMLFQ algorithm for multi-mobile robot path planning","authors":"Likun Hu, Chunyou Wei, Linfei Yin","doi":"10.1016/j.aei.2025.103398","DOIUrl":"10.1016/j.aei.2025.103398","url":null,"abstract":"<div><div>With the development of robotics, mobile robots (MRs) are widely applied in industrial and agricultural production. Reasonable path planning (PP) algorithms are the prerequisite for multi-mobile robot (MMR) systems to accomplish tasks. However, the existing PP algorithms of MMR systems still have the problems of being unable to dynamically assign tasks, not comprehensively considering the needs of kinematic constraints and dynamic obstacle avoidance, and poorly coordinating path conflicts. This study proposes a multi-agent proximal policy optimization-artificial potential field twin delayed deep deterministic policy gradient-improved multi-level feedback queue (MAPPO-ITD3-IMLFQ) algorithm for the PP of MMR systems. The proposed MAPPO-ITD3-IMLFQ algorithm combines the multi-agent proximal policy optimization (MAPPO) algorithm, the improved twin delayed deep deterministic policy gradient (ITD3) algorithm, and the improved multi-level feedback queue (IMLFQ) algorithm to form a PP algorithm for MMR system. The MRs apply the MAPPO algorithm to calculate task assignment (TA) schemes and provide sub-goal points for ITD3 algorithm. The MRs apply the ITD3 algorithm to calculate the path of the MRs. When the paths of different MRs conflict, the MR applies the IMLFQ algorithm to coordinate the movement of the MRs. The proposed MAPPO-ITD3-IMLFQ algorithm realizes the dynamic TA of the MMR system, meets the kinematic constraints and dynamic obstacle avoidance requirements of MRs, and coordinates path conflicts among the MRs. In this study, the proposed MAPPO-ITD3-IMLFQ algorithm is applied to different environments for the PP of MMRs. Experimental results show that: compared to the Hungarian algorithm and the genetic algorithm, the proposed MAPPO-ITD3-IMLFQ algorithm reduces the time spent on assigned tasks by 75.25 % and 77.44 %, respectively. Compared to the PP algorithms for reinforcement learning, the proposed MAPPO-ITD3-IMLFQ algorithm reduces the length of the planned path by 23.57 % on average.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103398"},"PeriodicalIF":8.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143887535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Lightweight segmentation model for automated facade installation in high-rise buildings 高层建筑立面自动化安装的轻量化分段模型

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-05-01 DOI: 10.1016/j.aei.2025.103374

Ahmed Yimam Hassen , Mehrdad Arashpour , Elahe Abdi

{"title":"Lightweight segmentation model for automated facade installation in high-rise buildings","authors":"Ahmed Yimam Hassen , Mehrdad Arashpour , Elahe Abdi","doi":"10.1016/j.aei.2025.103374","DOIUrl":"10.1016/j.aei.2025.103374","url":null,"abstract":"<div><div>The installation of curtain wall modules (CWM) in high-rise buildings is a complex task that poses significant safety risks due to manual labor, especially when working at great heights. Traditional methods are labor-intensive, time-consuming, and expose workers to hazards such as falls and equipment malfunctions. To mitigate these risks and enhance operational efficiency, automation and precise positioning of CWMs are essential. Accurate detection of installation locations becomes critical, as it enables crane operators or autonomous robots to position CWMs safely and precisely. This study introduces a novel approach utilizing semantic segmentation for detecting CWM installation locations. To address the challenges of deploying deep learning models on edge devices in construction environments, we propose Lightweight Attention Network (LANet), a lightweight, single-stream semantic segmentation architecture. LANet incorporates an optimized transformer module for global context modeling with linear complexity, enabling efficient feature extraction while maintaining computational efficiency. Additionally, we have developed a custom curtain wall dataset tailored for automating CWM installation, which was used to train and evaluate LANet. Experimental results demonstrate that LANet achieves competitive segmentation accuracy with only 1.92 million parameters, delivering real-time performance at 262 FPS on an RTX 3090 GPU and 19 FPS on a standard Intel i7 CPU. These results make LANet highly suitable for deployment in resource-constrained environments.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103374"},"PeriodicalIF":8.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143891487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0