Architectural intelligence最新文献

{"title":"Circular construction strategies of freeform structures based on smooth poly-hypar surfaces: design and fabrication of furniture-scale prototypes","authors":"Yuqian Liu,&nbsp;Ting Cao","doi":"10.1007/s44223-026-00136-4","DOIUrl":"10.1007/s44223-026-00136-4","url":null,"abstract":"<div><p>This paper introduces a method for freeform spatial structures based on planar node constraints for reusable components. The method generates diverse structural forms that satisfy planar node constraints based on the principles of smooth poly-hypar surfaces. Through systematic grouping of structural components according to disassembly and reassembly sequence, the method employs genetic algorithms to optimise component lengths within each group, enabling component reusability across different structural configurations. The planar node design enhances construction efficiency and reusability while reducing costs and improving structural aesthetics compared to conventional three-dimensional nodes. As a geometric and constructive proof-of-concept, the study validates the method’s feasibility through furniture-scale prototypes, offering new perspectives for sustainable architectural design through systematic component reuse.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00136-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149234714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EmboDrone: towards mobile proximal spatial intelligence for elderly outdoor assistance","authors":"Xiaomeng Zhang,&nbsp;Yu Yan,&nbsp;Jiadong Liang,&nbsp;Leiqing Xu","doi":"10.1007/s44223-026-00138-2","DOIUrl":"10.1007/s44223-026-00138-2","url":null,"abstract":"<div><p>Outdoor mobility presents persistent safety challenges for older adults, particularly in environments with uneven surfaces, dynamic traffic conditions, and dense crowds. Existing assistive technologies, ranging from wearables to walking companion robots and assistive drones, provide useful support but often operate within single scenarios, rely on command-driven behaviors, and lack legible interaction cues aligned with older adults’ cognitive tendencies. This study introduces EmboDrone, a near-body assistive drone designed to explore coherent, vision-based hazard perception across multiple outdoor situations. Using a Research-through-Design (RtD) approach, we first introduce Proximal Spatial Intelligence (PSI) as a conceptual framework for understanding how mobile assistance may coordinate environmental sensing, action responses, and user feedback within the user’s near-body space. Guided by this framework, we designed a mid-fidelity drone prototype that integrates RGB-D sensing and a YOLO-based visual recognition pipeline on an edge-computing flight platform, enabling automatic human-following and simple auditory alerts for potential hazards such as obstacles, traffic lights, and pedestrians. We then conducted a multi-stakeholder evaluation (<i>n</i> = 39) involving older adults, family caregivers, and experts to examine how different stakeholder roles perceive and respond to PSI-guided interaction behaviors. The findings provide preliminary insights into how proximal spatial intelligence is interpreted and trusted across stakeholder roles. Together, the PSI framework, the EmboDrone design exemplar, and the multi-stakeholder evaluation provide design directions for future mobile, proximity-based assistive systems for older adults in dynamic outdoor environments.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00138-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149035045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A simulation to design and voxel 3D printing (S-D-VP) workflow: multi-material integrated facade 2","authors":"Samuel Esses,&nbsp;Kostas Grigoriadis","doi":"10.1007/s44223-026-00135-5","DOIUrl":"10.1007/s44223-026-00135-5","url":null,"abstract":"<div><p>This paper presents the design and scale-model fabrication process for a multi-material building façade with embedded shading, structural performance, and ventilation voids. Despite increasing research on performance-driven façade systems, existing processes typically rely on mechanical assemblies and have yet to explore the incorporation of multiple façade functions within a single, materially continuous volume. In response, the study introduces a bespoke digital process that translates performance-informed design data into multi-material 3D-printing instructions compatible with PolyJet additive manufacturing. Digital analyses—specifically structural and computational fluid dynamics (CFD) simulations—inform the strategic placement of reinforcement thickenings and ventilation voids within the façade volume. Additional solar analyses inform the distribution of variable transparencies. These parameters are subsequently translated into material properties for voxel-based multi-material 3D-printing through a custom-built bitmap-generation slicer. The result is a simulation-to-design-to-voxel-3D-printing (S-D-VP) methodology demonstrated through a scaled architectural model, enabling architects and designers to convert analytical data into 3D-printed outputs with variable material properties.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00135-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149034820","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A knowledge-augmented diffusion framework for generating regionally distinctive architectural facades: a case study of Huizhou dwellings","authors":"Zhe Guo,&nbsp;Yuhang Kong,&nbsp;Zihuan Zhang,&nbsp;Qingyi Xiong","doi":"10.1007/s44223-026-00137-3","DOIUrl":"10.1007/s44223-026-00137-3","url":null,"abstract":"<div><p>Generating heritage facade images with recognizable regional style is increasingly feasible with diffusion models, but prompt engineering or standard fine-tuning alone rarely encodes the construction logic, spatial hierarchy, and symbolic rules that make such heritage architecture intelligible. This gap is critical for Huizhou traditional dwellings (HzTD), a distinctive type of vernacular architecture in southern Anhui Province, China, whose facade forms depend on typological relations among courtyard layouts, timber-frame bay rhythms, materials and decorative motifs. To address this gap, this study proposes a Knowledge-Augmented Stable Diffusion Model (KA-SDM) for semantic representation and controlled facade generation of HzTD. The framework first organizes construction knowledge into a CRM-inspired property graph schema implemented in Neo4j, where posts, spaces, ratios, wall plans, elements, and details are represented as queryable entities and relations. Rule-based graph inference then derives facade composition attributes, opening configurations, and detail semantics, which are converted by a Prompt-Conditioned Semantic Controller (PCSC) into structured prompt fragments and ControlNet masks. Experiments on Huizhou dwelling facade images from Xidi Village compare KA-SDM with a Stable Diffusion (SD) baseline and ablation settings based on LoRA and ControlNet. The results indicate that knowledge-guided semantic control improves typological consistency, structural plausibility, and expert-rated semantic alignment while preserving visual diversity. This research demonstrates a practical pathway for coupling architectural heritage knowledge representation with generative AI, supporting cultural documentation, design assistance, and virtual reconstruction.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00137-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148986237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Tool to teammate: a systematic review on human–robot collaboration in architectural fabrication","authors":"Tahmures Ghiyasi,&nbsp;Ali Ghazvinian","doi":"10.1007/s44223-026-00121-x","DOIUrl":"10.1007/s44223-026-00121-x","url":null,"abstract":"<div><p>The evolving field of Human–Robot Collaboration (HRC) in the Architecture, Engineering, Fabrication, and Construction (AEFC) academia and Industry requires new approaches that reconcile human cognitive adaptability with robotic precision to address the unpredictability of construction environments. Architectural fabrication operates under site-specific conditions, material irregularities, and design uniqueness that limit the effectiveness of conventional automation, especially in comparison to the controlled manufacturing settings. This review investigates how frameworks of HRC in architecture have evolved from tool-based automation toward shared and adaptive systems, and what challenges persist in translating these developments into real-world construction practices. A systematic review was conducted on 110 peer-reviewed studies, each analyzed across various analytical dimensions: level of collaboration, materials, fabrication scale, digital and physical methods. To organize this diversity, the review introduces a six-level classification of human–robot collaboration, starting from manual control to supervised autonomy, collaborative control, adaptive collaboration, fully autonomous systems, and finally collective human–robot collaboration. Each level builds upon and evolves from the insights of the preceding ones. This framework enables a consistent comparison of technological capabilities, degrees of autonomy, and modes of human involvement across studies. The synthesis indicates gradual evolution in HRC frameworks. The more basic levels retain strong connections to craft and material intuition but remain constrained in scalability and repeatability. Intermediate levels incorporate real-time feedback and learning-based responsiveness, enabling more balanced task sharing between humans and robots. The advanced levels demonstrate emerging capacities for distributed intelligence and coordinated multi-agent behavior, yet they encounter significant technical, operational, and economic limitations that restrict their deployment beyond experimental settings. Across all levels, advances in sensing, simulation, and human–machine interface design are steadily expanding the domain of robotic construction while revealing persistent gaps between research and practice. The trajectory of HRC in architecture represents a broader socio-technical transformation that redefines the relationship between design, labor, and automation. Future progress depends on integrating collaborative robotics into construction workflows, fostering democratization of technology, and establishing robust frameworks for safety, training, and regulation that support human–robot coexistence on site. Bridging the divide between experimental success and scalable application will require sustained interdisciplinary collaboration among architects, engineers, robotic experts, and social scientists to advance a more adaptive and human-centered construction future.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00121-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148782673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From shape grammars to machine learning: a plea for the neuro-symbolic approach to generative AI","authors":"Mario Carpo","doi":"10.1007/s44223-026-00129-3","DOIUrl":"10.1007/s44223-026-00129-3","url":null,"abstract":"<div><p>A comparison between examples of style transfer and stylistic imitation as envisaged by the theory of Shape Grammars in the 1970s and those enabled by Generative AI today (with case studies and student work from a research seminar). The essay emphasizes the limits of Generative AI as a pure imitation machine, and argues for a merger of old-school rule-based systems and contemporary LLM-based machine learning.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00129-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148751704","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multilevel computing framework for behavior recognition and interaction in mixed reality workspaces","authors":"Menghang Liu,&nbsp;Ruyi Yang,&nbsp;Luqi Feng,&nbsp;Yanzi Jin,&nbsp;Wenjie Wang,&nbsp;Chao Yan","doi":"10.1007/s44223-026-00127-5","DOIUrl":"10.1007/s44223-026-00127-5","url":null,"abstract":"<div><p>Mixed Reality (MR) technology, as an emerging interactive medium, has seen widespread application in various everyday fields in recent years, offering significant convenience for intelligent interactions between humans and spaces. However, current MR environments primarily rely on single-input interaction methods, where specific behaviors or body movements are recognized to drive changes in the virtual environment. Such approaches often overlook the comprehensive and interrelated nature of human behaviors, potentially diminishing the immersive user experience. This paper proposes an improved interaction model that aims to comprehensively perceive multi-scale human behaviors, including social, global-postural, local-gestural, and facial scales. Using a smart office space as the design prototype, the system integrates body tracking and affective computing techniques, and employs Grasshopper and Unity for real-time spatial modeling and rendering. This allows the MR environment to dynamically respond to user behaviors and support flexible spatial transformations. An experimental study provides initial evidence that the proposed framework can effectively sense user behaviors and dynamically adapt environmental changes. The results suggest potential improvements in spatial immersion and interaction naturalness, providing a reference for further exploration of behavior-driven interaction in intelligent MR environments.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00127-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148750947","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparative analysis of image control workflows in generative AI for architectural design: a FLUX.1-based study","authors":"Yang Yu,&nbsp;Ding Wen Bao,&nbsp;Yuqian Liu","doi":"10.1007/s44223-026-00134-6","DOIUrl":"10.1007/s44223-026-00134-6","url":null,"abstract":"<div><p>Text-to-image generative models often struggle to maintain visual consistency across outputs, which limits their applicability in professional architectural rendering workflows. This study examines three FLUX.1-based image-to-image workflows for architectural aerial renderings. These approaches include generation using prompts with Florence-2 providing descriptions, fine-tuning using LoRA on datasets specific to the domain, and FLUX.1 REDUX that uses image conditioning. Analysis using measures reveals that FLUX.1 REDUX shows the most consistency and maintains architectural details in the data, while generation using prompts shows high variation and fine-tuning using LoRA shows limited improvement in results. The results indicate that image-conditioned approaches provide stronger overall reference-image fidelity than text-only prompting and LoRA-based adaptation, offering a practical basis for more reliable generative workflows in architectural design.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00134-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148613704","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An empirical framework for sustainable building materials selection in residential construction: evidence from Tanzania","authors":"Constant Mahame,&nbsp;Geraldine J. Kikwasi,&nbsp;Modest M. Baruti,&nbsp;Olukanni O. David","doi":"10.1007/s44223-026-00132-8","DOIUrl":"10.1007/s44223-026-00132-8","url":null,"abstract":"<div><p>Across developing countries, the selection of building materials still relies on short-term costs and practitioners’ familiarity rather than structured sustainability assessments (SSAs), despite rapid urbanisation and rising housing demand. A prior systematic review identified 19 factors across environmental, economic, and socio-cultural dimensions; however, it did not establish their relative importance in any specific context or provide an operational tool to translate these factors into structured guidance for early-stage material evaluation. This study addresses both gaps in Tanzania, a rapidly urbanising developing country, through a structured questionnaire survey distributed to 326 registered construction professionals, yielding 156 usable responses from architects, engineers, and quantity surveyors with residential project experience. The reliability of the 19-factor instrument was confirmed using Cronbach’s alpha, which showed good to excellent internal consistency. Factor importance was evaluated using the Relative Importance Index (RII) within a triple bottom line framework. The results indicate that environmental factors dominate professional priorities. Material environmental impact, ecosystem protection, and waste management were the highest-ranked factors. Economic factors follow, with the life-cycle cost ranked above the initial cost, showing attention to the long-term value. Socio-cultural factors ranked lower overall, although health and safety remained important. These findings challenge the assumption that cost dominates decision-making in developing countries and align with the sustainability objectives of Tanzanian professionals. This study developed a weighted evaluation matrix that translates ranked factors into a practical tool for early stage material selection, where detailed data are often limited. Future work should apply multi-criteria methods, such as the Analytic Hierarchy Process, and validate the framework through case-based applications, including testing the transferability of these priorities to other developing-country residential construction sectors.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00132-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148613813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Responsive discrete deposition modelling for environment-situated additive manufacturing","authors":"Lucas Helle,&nbsp;Joseph Naguib,&nbsp;Roberto Naboni","doi":"10.1007/s44223-026-00133-7","DOIUrl":"10.1007/s44223-026-00133-7","url":null,"abstract":"<div><p>Conventional additive manufacturing relies on precomputed toolpaths and assumes stable material and environmental conditions, limiting its capacity to adapt during fabrication. This study introduces Responsive Discrete Deposition Modelling (R-DDM), a fabrication framework that replaces continuous extrusion with bead-scale deposition recalculated during fabrication through environmental sensing and agent-based decision-making. R-DDM formalises a closed sensing–response–action loop in which each material placement responds to live input, demonstrated here through light-seeking behaviours inspired by phototropism. A calibration study supports precise dosage control, while a parallel simulation environment enables behavioural definition and preliminary outcome exploration. Physical experiments show that branching, merging, and hybrid trajectories can be reliably produced without predefined geometry, and exhibit strong correspondence with simulation. By structuring fabrication as a sequence of local, stimulus-informed actions, R-DDM enables adaptive material organisation and offers a model for situated, environment-aware construction systems, particularly relevant for unstructured or variable contexts where standard toolpath logic fails.</p></div>","PeriodicalId":72270,"journal":{"name":"Architectural intelligence","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s44223-026-00133-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148613922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}