Automation in Construction最新文献

英文中文

Weakly-aligned cross-modal learning framework for subsurface defect segmentation on building façades using UAVs 基于弱对齐跨模态学习框架的无人机建筑表面缺陷分割

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105946

Sudao He , Gang Zhao , Jun Chen , Shenghan Zhang , Dhanada Mishra , Matthew Ming-Fai Yuen

Infrared (IR) thermography combined with Unmanned Aerial Vehicles (UAVs) offers an innovative approach for automated building façades inspections. However, extracting quantitative defect information from a single image poses a significant challenge. To address this, this paper introduces a Weakly-aligned Cross-modal Learning framework for subsurface defect segmentation using UAVs. This framework consists of two main components: the Multimodal Feature Description Network (MFDN) and the Prompt-aided Cross-modal Graph Learning (PCGL) algorithm. Initially, RGB–IR image pairs are processed by MFDN to extract feature descriptors for multi-modal alignment. The PCGL algorithm identifies visually critical areas through graph partitioning on a Wasserstein graph. These critical areas are transferred to the aligned IR image, and a Wasserstein Adjacency Graph (WAG) is constructed based on masked superpixel segmentation. Finally, the defects contours are pinpointed by detecting abnormal vertices of the WAG. The effectiveness is validated through controlled laboratory experiments and field applications on tiled façades.

红外（IR）热成像技术与无人机（uav）相结合，为自动化建筑立面检测提供了一种创新方法。然而，从单个图像中提取定量缺陷信息是一个重大挑战。为了解决这个问题，本文引入了一个弱对齐的跨模态学习框架，用于使用无人机进行地下缺陷分割。该框架由两个主要部分组成：多模态特征描述网络（MFDN）和快速辅助跨模态图学习（PCGL）算法。首先，对RGB-IR图像对进行MFDN处理，提取特征描述符，用于多模态对齐。PCGL算法通过在Wasserstein图上进行图划分来识别视觉上的关键区域。将这些关键区域转移到对齐后的红外图像上，并基于掩码超像素分割构建Wasserstein邻接图（WAG）。最后，通过检测WAG的异常顶点来确定缺陷轮廓。通过室内对照试验和现场应用，验证了该方法的有效性。

{"title":"Weakly-aligned cross-modal learning framework for subsurface defect segmentation on building façades using UAVs","authors":"Sudao He , Gang Zhao , Jun Chen , Shenghan Zhang , Dhanada Mishra , Matthew Ming-Fai Yuen","doi":"10.1016/j.autcon.2024.105946","DOIUrl":"10.1016/j.autcon.2024.105946","url":null,"abstract":"<div><div>Infrared (IR) thermography combined with Unmanned Aerial Vehicles (UAVs) offers an innovative approach for automated building façades inspections. However, extracting quantitative defect information from a single image poses a significant challenge. To address this, this paper introduces a Weakly-aligned Cross-modal Learning framework for subsurface defect segmentation using UAVs. This framework consists of two main components: the Multimodal Feature Description Network (MFDN) and the Prompt-aided Cross-modal Graph Learning (PCGL) algorithm. Initially, RGB–IR image pairs are processed by MFDN to extract feature descriptors for multi-modal alignment. The PCGL algorithm identifies visually critical areas through graph partitioning on a Wasserstein graph. These critical areas are transferred to the aligned IR image, and a Wasserstein Adjacency Graph (WAG) is constructed based on masked superpixel segmentation. Finally, the defects contours are pinpointed by detecting abnormal vertices of the WAG. The effectiveness is validated through controlled laboratory experiments and field applications on tiled façades.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"170 ","pages":"Article 105946"},"PeriodicalIF":9.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142939733","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

Self-training method for structural crack detection using image blending-based domain mixing and mutual learning 基于图像混合的区域混合和相互学习的结构裂纹检测自训练方法

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105892

Quang Du Nguyen , Huu-Tai Thai , Son Dong Nguyen

Deep learning-based structural crack detection utilizing fully supervised methods requires laborious labeling of training data. Moreover, models trained on one dataset often experience significant performance drops when applied to others due to domain shifts prompted by diverse structures, materials, and environmental conditions. This paper addresses the issues by introducing a robust self-training domain adaptive segmentation (STDASeg) pipeline. STDASeg incorporates an image blending-based domain mixing module to minimize domain discrepancies. Additionally, STDASeg involves a two-stage self-training framework characterized by the mutual learning scheme between Convolutional Neural Networks and Transformers, effectively learning domain invariant features from the two domains. Comprehensive evaluations across three challenging cross-dataset crack detection scenarios highlight the superiority of STDASeg over traditional supervised training approaches and current state-of-the-art methods. These results confirm the stability of STDASeg, thus supporting more efficient infrastructure assessments.

利用完全监督方法的基于深度学习的结构裂纹检测需要对训练数据进行费力的标记。此外，由于不同的结构、材料和环境条件引起的领域转移，在一个数据集上训练的模型在应用于其他数据集时往往会出现显著的性能下降。本文通过引入鲁棒自训练域自适应分割（STDASeg）管道来解决这些问题。STDASeg结合了一个基于图像混合的域混合模块，以最大限度地减少域差异。此外，STDASeg涉及一个两阶段的自训练框架，该框架以卷积神经网络和变压器之间的相互学习方案为特征，有效地从两个域中学习域不变特征。对三个具有挑战性的跨数据集裂缝检测场景的综合评估突出了STDASeg优于传统的监督训练方法和当前最先进的方法。这些结果证实了STDASeg的稳定性，从而支持更有效的基础设施评估。

{"title":"Self-training method for structural crack detection using image blending-based domain mixing and mutual learning","authors":"Quang Du Nguyen , Huu-Tai Thai , Son Dong Nguyen","doi":"10.1016/j.autcon.2024.105892","DOIUrl":"10.1016/j.autcon.2024.105892","url":null,"abstract":"<div><div>Deep learning-based structural crack detection utilizing fully supervised methods requires laborious labeling of training data. Moreover, models trained on one dataset often experience significant performance drops when applied to others due to domain shifts prompted by diverse structures, materials, and environmental conditions. This paper addresses the issues by introducing a robust self-training domain adaptive segmentation (STDASeg) pipeline. STDASeg incorporates an image blending-based domain mixing module to minimize domain discrepancies. Additionally, STDASeg involves a two-stage self-training framework characterized by the mutual learning scheme between Convolutional Neural Networks and Transformers, effectively learning domain invariant features from the two domains. Comprehensive evaluations across three challenging cross-dataset crack detection scenarios highlight the superiority of STDASeg over traditional supervised training approaches and current state-of-the-art methods. These results confirm the stability of STDASeg, thus supporting more efficient infrastructure assessments.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"170 ","pages":"Article 105892"},"PeriodicalIF":9.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142788866","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Automated six-degree-of-freedom Stewart platform for heavy floor tiling 自动六自由度Stewart平台，用于重型地砖

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105932

Siwei Chang , Zemin Lyu , Jinhua Chen , Tong Hu , Rui Feng , Haobo Liang

While existing floor tiling robots provide automated tiling for small tiles, robots designed for large and heavy tiles are rare. This paper develops a six-degree-of-freedom Stewart platform-based floor tiling robot for automated tiling of heavy tiles. The key contributions of this paper are: 1) establishing mechanical and kinematic models for a parallel robot to enhance the payload capacity of existing floor tiling robots. 2) designing a dual-camera system for precise visual alignment by capturing tile corner points from a complete perspective. Experimental validation demonstrated the robot's ability to automatically tile heavy floor tiles, with highly synchronized motions. The dual camera system achieved angle and distance deviations within ±0.001° and 0.5 mm. Quantitative analysis using the Borg RPE scale and EMG signals validated a reduction in physical strain. This research provides a feasible solution for automating heavy floor tile installation, effectively mitigating physical fatigue while enhancing the tiling alignment precision.

虽然现有的地板贴砖机器人可以自动贴小瓷砖，但为大型和重型瓷砖设计的机器人却很少见。本文研制了一种基于Stewart平台的六自由度地砖机器人，用于重型地砖的自动铺砖。本文的主要贡献有：1)建立了并联机器人的力学和运动学模型，以提高现有地砖机器人的有效载荷能力。2)设计双摄像头系统，从完整的角度捕捉瓷砖角点，实现精确的视觉对齐。实验验证表明，该机器人能够以高度同步的动作自动铺贴沉重的地砖。双摄像头系统实现了±0.001°和0.5 mm范围内的角度和距离偏差。使用Borg RPE量表和肌电图信号的定量分析证实了物理应变的减少。本研究为重型地砖安装自动化提供了一种可行的解决方案，有效地减轻了体力疲劳，同时提高了地砖对齐精度。

引用次数: 0

Structural design and fabrication of concrete reinforcement with layout optimisation and robotic filament winding 结构设计和制造与布局优化和机器人长丝缠绕混凝土钢筋

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105952

Robin Oval , John Orr , Paul Shepherd

Reinforced concrete is a major contributor to the environmental impact of the construction industry, due not only to its cement content, but also its steel tensile reinforcement, estimated to represent around 40% of the material embodied carbon. Reinforcement has a significant contribution because of construction rationalisation, resulting in regular cages of steel bars, despite the availability of structural-optimisation algorithms and additive-manufacturing technologies. This paper fuses computational design and digital fabrication, to optimise the reinforcement layout of concrete structures, by designing with constrained layout optimisation of strut-and-tie models where the ties are produced with robotic filament winding. The methodology is presented, implemented in open-source code, and illustrated on beam and plate reinforcement applications. The numerical studies yield a discussion about parameter selection and constraint influence on material and construction efficiency trade-offs. Small-scale physical prototypes up to 50 cm

\times

50 cm provide a proof-of-concept.

钢筋混凝土是建筑行业对环境影响的主要贡献者，不仅是因为它的水泥含量，还因为它的钢筋拉伸加固，估计占材料隐含碳的40%左右。尽管有结构优化算法和增材制造技术，但由于结构合理化，钢筋形成了规则的钢筋笼，因此加固起到了重要作用。本文将计算设计和数字制造相结合，通过约束布局优化设计钢筋混凝土结构模型，其中钢筋是由机器人丝缠绕生产的。本文给出了该方法，并在开源代码中实现，并在梁和板加固应用中进行了说明。数值研究讨论了参数选择和约束对材料和施工效率权衡的影响。高达50厘米× 50厘米的小规模物理原型提供了概念验证。

{"title":"Structural design and fabrication of concrete reinforcement with layout optimisation and robotic filament winding","authors":"Robin Oval , John Orr , Paul Shepherd","doi":"10.1016/j.autcon.2024.105952","DOIUrl":"10.1016/j.autcon.2024.105952","url":null,"abstract":"<div><div>Reinforced concrete is a major contributor to the environmental impact of the construction industry, due not only to its cement content, but also its steel tensile reinforcement, estimated to represent around 40% of the material embodied carbon. Reinforcement has a significant contribution because of construction rationalisation, resulting in regular cages of steel bars, despite the availability of structural-optimisation algorithms and additive-manufacturing technologies. This paper fuses computational design and digital fabrication, to optimise the reinforcement layout of concrete structures, by designing with constrained layout optimisation of strut-and-tie models where the ties are produced with robotic filament winding. The methodology is presented, implemented in open-source code, and illustrated on beam and plate reinforcement applications. The numerical studies yield a discussion about parameter selection and constraint influence on material and construction efficiency trade-offs. Small-scale physical prototypes up to 50 cm <span><math><mo>×</mo></math></span> 50 cm provide a proof-of-concept.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"170 ","pages":"Article 105952"},"PeriodicalIF":9.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142939669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Digital twin construction with a focus on human twin interfaces 数字孪生体构建，重点关注孪生体人机界面

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105924

Ranjith K. Soman , Karim Farghaly , Grant Mills , Jennifer Whyte

Despite the growing emphasis on digital twins in construction, there is limited understanding of how to enable effective human interaction with these systems, limiting their potential to augment decision-making. This paper investigates the research question: “How can construction control rooms be utilized as digital twin interfaces to enhance the accuracy and efficiency of decision-making in the digital twin construction workflow?”. Design science research was used to develop a framework for human-digital twin interfaces, and it was evaluated in a real-world construction project. Findings reveal that control rooms can serve as dynamic interfaces within the digital twin ecosystem, improving coordination efficiency and decision-making accuracy. This finding is significant for practitioners and researchers, as it highlights the role of digital twin interfaces in augmenting decision-making. The paper opens avenues for future studies of human-digital twin interaction and machine learning in construction, such as imitation learning, codifying tacit knowledge, and new HCI paradigms.

尽管在建设中越来越重视数字孪生，但人们对如何使人类与这些系统有效互动的理解有限，限制了它们增强决策的潜力。本文探讨了“如何利用施工控制室作为数字孪生接口，提高数字孪生施工工作流程决策的准确性和效率”这一研究问题。利用设计科学研究开发了人-数字孪生界面框架，并在实际工程中进行了评估。研究结果表明，控制室可以作为数字孪生生态系统中的动态接口，提高协调效率和决策准确性。这一发现对从业者和研究人员来说意义重大，因为它强调了数字孪生接口在增强决策方面的作用。本文为未来人类-数字孪生交互和机器学习在建筑中的研究开辟了道路，如模仿学习、编码隐性知识和新的HCI范式。

引用次数: 0

Automated reality capture for indoor inspection using BIM and a multi-sensor quadruped robot 利用 BIM 和多传感器四足机器人进行室内检测的自动现实捕捉

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105930

Zhengyi Chen , Changhao Song , Boyu Wang , Xingyu Tao , Xiao Zhang , Fangzhou Lin , Jack C.P. Cheng

This paper presents a real-time, cost-effective navigation and localization framework tailored for quadruped robot-based indoor inspections. A 4D Building Information Model is utilized to generate a navigation map, supporting robotic pose initialization and path planning. The framework integrates a cost-effective, multi-sensor SLAM system that combines inertial-corrected 2D laser scans with fused laser and visual-inertial SLAM. Additionally, a deep-learning-based object recognition model is trained for multi-dimensional reality capture, enhancing comprehensive indoor element inspection. Validated on a quadruped robot equipped with an RGB-D camera, IMU, and 2D LiDAR in an academic setting, the framework achieved collision-free navigation, reduced localization drift by 71.77 % compared to traditional SLAM methods, and provided accurate large-scale point cloud reconstruction with 0.119-m precision. Furthermore, the object detection model attained mean average precision scores of 73.7 % for 2D detection and 62.9 % for 3D detection.

本文提出了一种为基于四足机器人的室内检测量身定制的实时、经济高效的导航和定位框架。利用4D建筑信息模型生成导航地图，支持机器人姿态初始化和路径规划。该框架集成了一个具有成本效益的多传感器SLAM系统，将惯性校正的2D激光扫描与融合激光和视觉惯性SLAM相结合。此外，还训练了基于深度学习的物体识别模型，用于多维现实捕获，增强室内元素的综合检测。在一个配备RGB-D相机、IMU和2D LiDAR的四足机器人上进行了学术环境验证，该框架实现了无碰撞导航，与传统SLAM方法相比，定位漂移减少了71.77%，并提供了精度为0.119 m的精确大尺度点云重建。此外，目标检测模型在二维检测和三维检测方面的平均精度分别达到73.7%和62.9%。

{"title":"Automated reality capture for indoor inspection using BIM and a multi-sensor quadruped robot","authors":"Zhengyi Chen , Changhao Song , Boyu Wang , Xingyu Tao , Xiao Zhang , Fangzhou Lin , Jack C.P. Cheng","doi":"10.1016/j.autcon.2024.105930","DOIUrl":"10.1016/j.autcon.2024.105930","url":null,"abstract":"<div><div>This paper presents a real-time, cost-effective navigation and localization framework tailored for quadruped robot-based indoor inspections. A 4D Building Information Model is utilized to generate a navigation map, supporting robotic pose initialization and path planning. The framework integrates a cost-effective, multi-sensor SLAM system that combines inertial-corrected 2D laser scans with fused laser and visual-inertial SLAM. Additionally, a deep-learning-based object recognition model is trained for multi-dimensional reality capture, enhancing comprehensive indoor element inspection. Validated on a quadruped robot equipped with an RGB-D camera, IMU, and 2D LiDAR in an academic setting, the framework achieved collision-free navigation, reduced localization drift by 71.77 % compared to traditional SLAM methods, and provided accurate large-scale point cloud reconstruction with 0.119-m precision. Furthermore, the object detection model attained mean average precision scores of 73.7 % for 2D detection and 62.9 % for 3D detection.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"170 ","pages":"Article 105930"},"PeriodicalIF":9.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142867664","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

Semantic navigation for automated robotic inspection and indoor environment quality monitoring 用于自动化机器人检测和室内环境质量监测的语义导航

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105949

Difeng Hu, Vincent J.L. Gan

Maintaining a comfortable indoor environment is essential for enhancing occupant well-being. However, traditional inspection methods rely on manual input of precise coordinates for target objects, limiting efficiency. This paper proposes a semantic navigation approach to improve robotic inspection intelligence and efficiency. A revised RandLA-Net and KNN algorithm construct a semantic map rich in detailed object information, supporting semantic navigation. Subsequently, an object instance reasoning algorithm automatically identifies and extracts target object coordinates from the semantic map using human-like language commands. Given the position information, a semantics-aware A* algorithm calculates safer, more efficient navigation paths through enhanced robot-environment interaction. Experiments demonstrate a position accuracy of ∼0.08 m for objects in the semantic map and effective coordinate extraction by the reasoning algorithm. The semantics-aware A* algorithm generates paths farther from obstacles and cluttered areas with less computational time, indicating its superior performance in terms of the robot's safety and inspection efficiency.

保持舒适的室内环境对于提高居住者的幸福感至关重要。然而，传统的检测方法依赖于人工输入目标物体的精确坐标，限制了效率。为了提高机器人检测的智能和效率，提出了一种语义导航方法。改进的RandLA-Net和KNN算法构建了一个包含丰富详细目标信息的语义地图，支持语义导航。随后，对象实例推理算法使用类人语言命令从语义图中自动识别和提取目标对象坐标。给定位置信息，语义感知的a *算法通过增强机器人与环境的交互计算出更安全、更有效的导航路径。实验表明，该推理算法在语义图中对目标的定位精度为~ 0.08 m，并能有效地提取坐标。语义感知的A*算法以更少的计算时间生成了远离障碍物和杂乱区域的路径，表明其在机器人的安全性和检测效率方面具有优越的性能。

{"title":"Semantic navigation for automated robotic inspection and indoor environment quality monitoring","authors":"Difeng Hu, Vincent J.L. Gan","doi":"10.1016/j.autcon.2024.105949","DOIUrl":"10.1016/j.autcon.2024.105949","url":null,"abstract":"<div><div>Maintaining a comfortable indoor environment is essential for enhancing occupant well-being. However, traditional inspection methods rely on manual input of precise coordinates for target objects, limiting efficiency. This paper proposes a semantic navigation approach to improve robotic inspection intelligence and efficiency. A revised RandLA-Net and KNN algorithm construct a semantic map rich in detailed object information, supporting semantic navigation. Subsequently, an object instance reasoning algorithm automatically identifies and extracts target object coordinates from the semantic map using human-like language commands. Given the position information, a semantics-aware A* algorithm calculates safer, more efficient navigation paths through enhanced robot-environment interaction. Experiments demonstrate a position accuracy of ∼0.08 m for objects in the semantic map and effective coordinate extraction by the reasoning algorithm. The semantics-aware A* algorithm generates paths farther from obstacles and cluttered areas with less computational time, indicating its superior performance in terms of the robot's safety and inspection efficiency.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"170 ","pages":"Article 105949"},"PeriodicalIF":9.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142888278","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

Hybrid deep learning model for accurate cost and schedule estimation in construction projects using sequential and non-sequential data 利用顺序和非顺序数据的混合深度学习模型，准确估算建筑项目的成本和进度

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105904

Min-Yuan Cheng, Quoc-Tuan Vu, Frederik Elly Gosal

Accurate estimation of construction costs and schedules is crucial for optimizing project planning and resource allocation. Most current approaches utilize traditional statistical analysis and machine learning techniques to process the vast amounts of data regularly generated in construction environments. However, these approaches do not adequately capture the intricate patterns in either time-dependent or time-independent data. Thus, a hybrid deep learning model (NN-BiGRU), combining Neural Network (NN) for time-independent and Bidirectional Gated Recurrent Unit (BiGRU) for time-dependent, was developed in this paper to estimate the final cost and schedule to completion of projects. The Optical Microscope Algorithm (OMA) was used to fine-tune the NN-BiGRU model (OMA-NN-BiGRU). The proposed model earned Reference Index (RI) values of 0.977 for construction costs and 0.932 for completion schedules. These findings underscore the potential of the OMA-NN-BiGRU model to provide highly accurate predictions, enabling stakeholders to make informed decisions that promote project efficiency and overall success.

准确估算建设成本和进度对于优化项目规划和资源配置至关重要。目前大多数方法利用传统的统计分析和机器学习技术来处理建筑环境中定期生成的大量数据。然而，这些方法并不能充分捕捉到依赖时间或不依赖时间的数据中的复杂模式。因此，本文开发了一种混合深度学习模型（NN-BiGRU），结合了时间无关的神经网络（NN）和时间相关的双向门控循环单元（BiGRU），以估计项目的最终成本和完成进度。采用光学显微镜算法（OMA）对NN-BiGRU模型（OMA-NN-BiGRU）进行微调。该模型的建造成本参考指数（RI）为0.977，完工进度参考指数（RI）为0.932。这些发现强调了OMA-NN-BiGRU模型提供高度准确预测的潜力，使利益相关者能够做出明智的决策，从而提高项目效率和整体成功。

{"title":"Hybrid deep learning model for accurate cost and schedule estimation in construction projects using sequential and non-sequential data","authors":"Min-Yuan Cheng, Quoc-Tuan Vu, Frederik Elly Gosal","doi":"10.1016/j.autcon.2024.105904","DOIUrl":"10.1016/j.autcon.2024.105904","url":null,"abstract":"<div><div>Accurate estimation of construction costs and schedules is crucial for optimizing project planning and resource allocation. Most current approaches utilize traditional statistical analysis and machine learning techniques to process the vast amounts of data regularly generated in construction environments. However, these approaches do not adequately capture the intricate patterns in either time-dependent or time-independent data. Thus, a hybrid deep learning model (NN-BiGRU), combining Neural Network (NN) for time-independent and Bidirectional Gated Recurrent Unit (BiGRU) for time-dependent, was developed in this paper to estimate the final cost and schedule to completion of projects. The Optical Microscope Algorithm (OMA) was used to fine-tune the NN-BiGRU model (OMA-NN-BiGRU). The proposed model earned Reference Index (RI) values of 0.977 for construction costs and 0.932 for completion schedules. These findings underscore the potential of the OMA-NN-BiGRU model to provide highly accurate predictions, enabling stakeholders to make informed decisions that promote project efficiency and overall success.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"170 ","pages":"Article 105904"},"PeriodicalIF":9.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142816520","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

Sensor adoption in the construction industry: Barriers, opportunities, and strategies 建筑行业采用传感器的情况：障碍、机遇和战略

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105937

Zhong Wang, Vicente A. González, Qipei Mei, Gaang Lee

This paper examines the underutilization of sensors in the construction industry despite their significant potential for improving performance. A systematic review was conducted on research published between 2004 and 2024, identifying 11 key barriers such as the need for advanced skill sets and user-centric design, lack of standardized practices, and challenges in data networks and management. The study applied both quantitative descriptive analysis and qualitative content analysis to explore these barriers across five stages of sensor adoption. A total of 63 articles were thoroughly reviewed to identify thematic patterns and chronological trends. The findings highlight critical areas that require attention, including the development of standardized protocols, enhancing data-driven decision-making with advanced analytics, and fostering industry-wide training programs. Additionally, leveraging Lean Construction 4.0 principles is proposed to address these challenges. The insights from this research aim to support the construction industry in integrating sensor technologies more effectively, leading to greater efficiency and improved performance.

尽管传感器在提高性能方面具有巨大潜力，但本文探讨了传感器在建筑行业利用不足的问题。本文对 2004 年至 2024 年间发表的研究进行了系统回顾，确定了 11 个关键障碍，如需要先进的技能组合和以用户为中心的设计、缺乏标准化实践以及数据网络和管理方面的挑战。研究采用定量描述性分析和定性内容分析的方法，探讨了传感器应用五个阶段中的这些障碍。共对 63 篇文章进行了全面审查，以确定主题模式和时间趋势。研究结果强调了需要关注的关键领域，包括制定标准化协议、利用先进的分析技术加强数据驱动决策以及促进全行业的培训计划。此外，还建议利用精益建造 4.0 原则来应对这些挑战。本研究的见解旨在支持建筑行业更有效地整合传感器技术，从而提高效率并改善性能。

{"title":"Sensor adoption in the construction industry: Barriers, opportunities, and strategies","authors":"Zhong Wang, Vicente A. González, Qipei Mei, Gaang Lee","doi":"10.1016/j.autcon.2024.105937","DOIUrl":"10.1016/j.autcon.2024.105937","url":null,"abstract":"<div><div>This paper examines the underutilization of sensors in the construction industry despite their significant potential for improving performance. A systematic review was conducted on research published between 2004 and 2024, identifying 11 key barriers such as the need for advanced skill sets and user-centric design, lack of standardized practices, and challenges in data networks and management. The study applied both quantitative descriptive analysis and qualitative content analysis to explore these barriers across five stages of sensor adoption. A total of 63 articles were thoroughly reviewed to identify thematic patterns and chronological trends. The findings highlight critical areas that require attention, including the development of standardized protocols, enhancing data-driven decision-making with advanced analytics, and fostering industry-wide training programs. Additionally, leveraging Lean Construction 4.0 principles is proposed to address these challenges. The insights from this research aim to support the construction industry in integrating sensor technologies more effectively, leading to greater efficiency and improved performance.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"170 ","pages":"Article 105937"},"PeriodicalIF":9.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142867674","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Real-time and high-accuracy defect monitoring for 3D concrete printing using transformer networks 基于变压器网络的三维混凝土打印缺陷实时、高精度监测

IF 9.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-02-01 DOI: 10.1016/j.autcon.2024.105925

Hongyu Zhao , Junbo Sun , Xiangyu Wang , Yufei Wang , Yang Su , Jun Wang , Li Wang

Defects and anomalies during the 3D concrete printing (3DCP) process significantly affect final construction quality. This paper proposes a real-time, high-accuracy method for monitoring defects in the printing process using a transformer-based detector. Despite limited data availability, deep learning-based data augmentation and image processing techniques were employed to enable effective training of this complex transformer model. A range of enhancement strategies was applied to the RT-DETR, resulting in remarkable improvements, including a mAP50 of 98.1 %, mAP50–95 of 68.0 %, and a computation speed of 72 FPS. The enhanced RT-DETR outperformed state-of-the-art detectors such as YOLOv8 and YOLOv7 in detecting defects in 3DCP. Furthermore, the improved RT-DETR was used to analyze the relationships between defect count, size, and printer parameters, providing guidance for operators to fine-tune printer settings and promptly address defects. This monitoring method reduces material waste and minimizes the risk of structural collapse during the printing process.

三维混凝土打印（3DCP）过程中的缺陷和异常会严重影响最终的建筑质量。本文提出了一种实时、高精度的方法，利用基于变压器的检测器监测打印过程中的缺陷。尽管数据可用性有限，但还是采用了基于深度学习的数据增强和图像处理技术，以便对这一复杂的变压器模型进行有效训练。对 RT-DETR 采用了一系列增强策略，取得了显著的改进，包括 mAP50 为 98.1%，mAP50-95 为 68.0%，计算速度为 72 FPS。增强型 RT-DETR 在检测 3DCP 中的缺陷方面优于 YOLOv8 和 YOLOv7 等最先进的检测器。此外，改进型 RT-DETR 还用于分析缺陷数量、大小和打印机参数之间的关系，为操作员微调打印机设置和及时处理缺陷提供指导。这种监测方法减少了材料浪费，并将印刷过程中结构坍塌的风险降至最低。

{"title":"Real-time and high-accuracy defect monitoring for 3D concrete printing using transformer networks","authors":"Hongyu Zhao , Junbo Sun , Xiangyu Wang , Yufei Wang , Yang Su , Jun Wang , Li Wang","doi":"10.1016/j.autcon.2024.105925","DOIUrl":"10.1016/j.autcon.2024.105925","url":null,"abstract":"<div><div>Defects and anomalies during the 3D concrete printing (3DCP) process significantly affect final construction quality. This paper proposes a real-time, high-accuracy method for monitoring defects in the printing process using a transformer-based detector. Despite limited data availability, deep learning-based data augmentation and image processing techniques were employed to enable effective training of this complex transformer model. A range of enhancement strategies was applied to the RT-DETR, resulting in remarkable improvements, including a mAP50 of 98.1 %, mAP50–95 of 68.0 %, and a computation speed of 72 FPS. The enhanced RT-DETR outperformed state-of-the-art detectors such as YOLOv8 and YOLOv7 in detecting defects in 3DCP. Furthermore, the improved RT-DETR was used to analyze the relationships between defect count, size, and printer parameters, providing guidance for operators to fine-tune printer settings and promptly address defects. This monitoring method reduces material waste and minimizes the risk of structural collapse during the printing process.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"170 ","pages":"Article 105925"},"PeriodicalIF":9.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142867683","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

首页上一页

下一页尾页

类型

全部化学•材料生命科学医学物理工程技术环境•农林材料科学地球科学法学管理学化学环境科学与生态学计算机科学教育学经济学农林科学人文科学生物学数学物理与天体物理心理学综合性期刊其他工业工程理学历史学农学文学信息工程

数据库

全部 ACS Publications Elsevier ieeexplore Springer The Royal Society of Chemistry Wiley

期刊

Automation in Construction

全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.

﹀