Symbiotic Intelligence

This study explores the integration of Topology Optimization (TO) with non-planar 3D concrete printing (NP-3DCP) to address environmental and material efficiency challenges in construction. We present a novel approach leveraging robotic NP-3DCP for creating lightweight, structurally optimized architectural components, specifically focusing on load-bearing walls. The advantages of this approach were demonstrated through the manufacturing of two prototypes, Shell Wall and Branch Wall, showcasing significant material savings and reduced carbon footprint without compromising structural integrity. Our methodology encompasses the generation of a non-planar printpath, material extrusion control for variable layer height, and the integration of rebar and thermal insulation within the casting process. The results showcase the potential of this approach in producing complex geometries with improved environmental performance, suggesting a promising direction for the future of sustainable concrete construction.

This study introduces a digital manufacturing approach for mini and micro-housing design, deploying 3D concrete printing. Combining shape grammar with robotic construction methods, it aims to revolutionize architectural practice by enabling mass customization while ensuring creativity and feasibility. Traditional construction methods often hinder affordability and design diversity, necessitating innovative approaches. Shape grammars, rooted in design language principles, facilitate architectural design exploration. This research focuses on developing a novel generative system and harnessing automated printing for efficiency, reduction of embodied carbon and waste reduction. The study proposes leveraging digital manufacturing and 3D concrete printing. Its methodology involves developing a design grammar, a computer implementation, and fabricating prototypes. The approach, currently undergoing physical prototyping, demonstrates customizable housing solutions, advocating for a streamlined approach to housing design and construction while addressing affordability and customization.

As a critical element in urban sustainability efforts, green roofs (GRs) can reduce building energy use and enhance ecological performance via additional growing medium and plant layers. However, the extra building components result in increased load and structural thickness as well as construction complexity. This paper proposes a design-fabrication method for a thin-vaulted green roof prototype with a compressive-only surface and upstand ribs along unevenly distributed stress lines for a lightweight and material-efficient structure. The structural efficiency of the proposed roof has been proven high compared to conventional flat roof through simulation. To realize such a non-standard and multifunctional structure, a hybrid formwork system is presented to deal with geometric complexity and functional integration by treating stay-in-place formwork as a function part. In particular, 3D clay printing (3DCP) is selected as eco-friendly formwork staying in the GRs to integrate with plant growth substrate, leading to a more simplified and sustainable fabrication. An empirical construction experiment is conducted to validate the proposed method on a 1:5 scale.

This paper introduces a positioning method based on the robot arm and laser-camera method, which is mainly oriented to the on-site additive construction and blockwork using mobile robots, and ensures the positioning accuracy of about 5 mm. This method provides a location scheme that is cheaper than directly using total stations and has higher accuracy than LiDAR or VSLAM. In addition, the method overcomes the limitation of the arm span on the positioning process: it is usually located by scanning the ArUco marker on the site by the camera at the end of the robot arm, which means that intensive positioning markers are required. The main equipment involved in the positioning process includes a laser light mounted at the end of the robot and a laser receiving panel as a marking point. The receiving panel is observed by an RGB camera to ensure that the light spot is projected in the center of the panel. The whole process is as follows: in advance, a total station is used to arrange several receiving panels at an interval of 5 to 10 m on the site, and then the mobile robot will shoot the laser at nearby receiving panels in turn, and push back its position through each axis angle to complete the positioning. The accuracy of this method mainly depends on the accuracy of the 4–6 axis of the robot. Therefore, this paper also introduces a calibration method to measure the mechanical error of the robot arm itself, to improve the positioning accuracy as much as possible. With the help of this method, many construction processes will achieve mobile construction and can bring some innovation to the construction process.

This research project presents an alternative approach to addressing the complex challenges of sustainability of the coastlines by integrating advanced technology solutions with ecological conservation principles. The paper introduces the Ecoblox, a modular infrastructure system consisting of interlocking blocks devised for attachment to seawalls to improve marine biodiversity at the water edge. The design of the Ecoblox system employs environmental data, data analytics, AI-powered generative algorithms, and digital fabrication to produce blocks with complex shapes and textures suitable for bio-marine habitats.The project is executed in two phases. This paper describes the initial phase, encompassing the prototyping process, construction, testing, and analysis of various Ecoblox versions. The primary objective of this phase is to assess multiple designs and evaluate their effects on biodiversity. Building upon the insights gained from the initial phase, Phase II of the study focuses on developing data-driven strategies and applying robotic 3D printing to refine the system’s design and construction.

This study proposes a novel solution for addressing coastal erosion and ecosystem degradation worldwide. By attaching a specially designed prosthesis to the curved surface of tetrapods, common marine structures in Korea, we aim to restore vital seaweed forest ecosystems within coastal areas. Leveraging 3D scanning, printing technologies, and robotic arms, we develop a customized attachment to seamlessly integrate with tetrapods, overcoming construction area limitations. Our approach utilizes Fused Granular Fabrication, employing biodegradable plastic made from recycled seaweed waste, promoting sustainable material use and disposal in marine environments. Method of attaching to the surface of the tetrapod was divided into three methods: Foldable 3D printing, On-surface 3D printing, and Barnacle 3D printing, which improves the two preceding methods. This innovative solution offers promise for sustainable coastal management and ecosystem restoration.

Contemporary building envelopes primarily rely on repetitive elements that fulfill mainly a singular purpose - a barrier between interior and exterior spaces. Implementing building envelopes featuring intricate geometries and non-repetitive tiles can significantly enhance the environmental performance of the structure. However, the current manufacturing processes for the required moulds are plagued by high costs, time consumption, labor-intensiveness, and mainly by using non-recyclable moulds. To address these challenges, the paper presents an innovative solution for designing and fabricating building envelope tiles with complex geometries by employing stay-in-place 3D knitted moulds. The moulds are digitally fabricated using innovative knitting procedures implemented on an industrial flat double-bed machine. The paper presents preliminary research outcomes, including a new digital design methodology for creating knitted moulds and a new fabrication method for buildings’ envelope tiles.

Augmented Reality (AR) allows workers to construct buildings accurately and intuitively without the need for traditional tools like 2-D drawings and rulers. However, accurately tracking worker’s pose remains a significant challenge in existing experiments due to their continuous and irregular movement. This research discusses a series of methods using cameras and algorithms to achieve the 6-DoF pose tracking function and reveal the relationship between each method and corresponding tracking accuracy in order to figure out a robust approach of AR-assisted assembly. This paper begins with a consideration of the possible limitations of existing methods including the image drift associated with visual SLAM and the time-consuming nature of fiducial markers. Next, the entire hardware and software framework was introduced, which elaborates on how the motion capture system is integrated into the AR-assisted assembly system. Then, some experiments have been carried out to demonstrate the connection between the system set up and pose tracking accuracy. This research shows the possibility to easily finish assembly task based on AR technology by integrating motion capture system.

This paper explores the design of standardized 3D-printed joints for small-scale architectural models, using the traditional mortise and tenon joint as a prototype. In comparison to traditional subtractive manufacturing for model production, 3D printing proves beneficial in saving time and materials. Therefore, many architectural students and design firms frequently use 3D printing for making models. Considering the limited size of 3D printers, convenience of transportation, and flexibility in demonstration, it is more advantageous to print model components with assembly joints rather than printing the entire model directly. This paper focuses on small-scale architectural models ranging from 1:50 to 1:200, conducting a typification, standardization and parameterization study of 3D-printed assembly joints. 13 standard joints are designed, forming a database and an interactive application.

Interactive bricklaying combines augmented reality technology with robot fabrication technology, utilizing both human intelligence and robot precision. Humans intuitively complete various differentiated bricklaying designs, while robot systems accurately execute bricklaying tasks. Traditionally, in such workflows, the design and fabrication processes are separated, with a lower level of human-robot interaction. By comparing with traditional interactive workflows through experimental studies, this research explores a higher degree of interactive workflow where design and fabrication alternate. It analyzes the impact, limitations, and potential of human involvement in linear robot fabrication processes. Using this new workflow, the study achieved the design and fabrication of a conical structure without prior design. During construction, the structure was gradually reduced in layers from the bottom up, with the number of bricks reduced and brick positions adjusted through manual observation, successfully controlling the tapering form and achieving closure of the circular brick wall. The new workflow utilizes human judgment to continuously determine the next design based on the current real construction situation, ensuring the rationality of the design and the stability of the construction process.

Moss Columns are a series of prototypes to examine how to combine living organisms with architecture. A renewed perspective on the built environment has increased in the post-COVID-19 era. Even various methods have been employed for sustainable design, the construction material itself in contemporary architecture have remained largely unchanged. In these two different types of experiment, I am presenting a direct embedding approach of plants into the artificial materials. Mosses, chosen as the primary plant for these experiments, are suitable due to their non-vascular nature, which means they do not grow tall like other plants and use roots solely for anchoring rather than nourishment. To investigate high-res and complicated pattern for embedding, geometry is manipulated and generated through computational design tools. Advanced construction technologies are employed to realize the complex forms: such as two different types of large-scale 3D printers with an industrial robotic arm.

This research presents a humidity-responsive facade fabricated using advanced techniques such as robotic 3D printing, CNC milled wood, and kerf application. Inspired by nature’s adaptability, exemplified by how pinecones respond to humidity levels, the concept of ‘animate materials’ like wood, which can adapt and respond to their environment, was explored. A circular manufacturing approach was adopted, utilizing CNC milling and robotic 3D printing. Sawdust, a by-product, was transformed with Chitosan-based hydrogel to create a 3D printable material. This material was laid down in three distinct layers, each serving a specific function and forming an active gradient layer that allows for kinetic motion. A material library was established to customize designs, aiming to create spaces that interact with local ecosystems and respond to environmental conditions. This approach challenges the conventional timber industry by introducing innovative materials for construction.

This study aims to optimize the additive manufacturing of chitosan/cellulose materials, enhancing their applicability. According to the United Nations Environment Programme (UNEP) report, 40% of global greenhouse gas emissions come from the construction industry. Our goal is to reduce emissions in construction, improve material efficiency, and lower emission intensity. The traditional linear economic model in construction leads to significant CO2 emissions and construction waste. Therefore, based on the principles of a circular economy, particularly in 3D printing technology, we investigated chitosan and cellulose as biodegradable materials, which are natural high-molecular-weight polymers. Past research has identified shrinkage issues in 3D printing using chitosan and cellulose. By optimizing the printing paths to increase the material’s contact area with air, we aimed to reduce overall structural shrinkage, enhancing the application scenarios of this novel material in construction. The research covers three dimensions: materials, software, and hardware. Experimental results demonstrate that the addition of paper fibers not only reduces shrinkage and cracking but also improves strength, making the material more stable. We utilized Rhino-Grasshopper software for analysis and path generation and integrated robots and extrusion devices in 3D printing technology. Through the implementation of the material-design-build process, we aim to enhance the prospects of degradable materials in the construction industry, mitigate the negative impacts on the building lifecycle, and promote a more environmentally friendly, conscious, and energy-efficient construction sector.

This paper introduces a 3D-printed façade module that cultivates Chlorella with CO2 to purify indoor air for inhabitants and supply biomass. Algae produces oxygen and biomass and enhances carbon sequestration through photosynthetic processes. The idea of integrating algae into architecture as bioreactors has been developed in recent years. In this research, the façade module consists of a hybrid framework and an algae cultivation apparatus. The hybrid framework is composed of aluminum profiles, a 3D-printed skin, slim solar panels for the pump battery, and fasteners. The algae culture system regulates the photosynthesis inside the tubular liquid to interact with the indoor air. The module’s freeform skin with grooves is 3D printed with large-scale Fused Granulate Fabrication (FGF). The convoluted tube along the grooves is always ascendant to make the air travel slowly but smoothly from bottom to top. The prototype installed on the building façade enables the algae organisms in the bioreactor to grow stably. This project introduces biochemistry processes into sustainable design toward metabolism in the built environment.

Over the last two decades, various contributions have shown how the use of graphic statics made it possible to design remarkable engineering structures. Vector-based Graphic Statics (VGS) has been presented elsewhere extensively as a method and plug-in for Grasshopper. This contribution shows an application to the design of a timber pavilion based on Graphic statics and its benefits for the use of materials.

This paper undertakes an in-depth investigation into a hybrid digital and analogue design process employed in crafting sand-based panel systems conducive to computational optimization. Tailored for extreme desert environments abundant in dune sand, these systems leverage the material’s inherent self-organizing properties. The methodology employs a multi-objective computational system strategically and culminates in an optimized, panelized architectural system. The approach emphasizes self-organization principles, initiated with physical experiments on natural dune sand piles. Advancing to controlled sand deposition on laser-cut planes facilitates precise configurations. The study systematically explores variables such as opening size and quantity, securing configurations with a binder and integrating them into diverse physical surface sequences. Computational analysis refines these sand patterns, identifying optimal configurations aligned with desert-specific contexts. This amalgamation of computational analysis and material processes enriches discussions on designing for extreme environments, aligning seamlessly with UN sustainability goals focused on sustainable communities, climate resilience, and responsible resource utilization.

In the context of global energy conservation and emission reduction, it is imperative to improve energy efficiency and promote sustainable development. The industry sector has consistently been a pivotal industry in implementing low-carbon practices, particularly in large industrial zones characterized by high energy consumption and carbon emissions. A focal point of research in such contexts revolves around the configuration of energy systems in industrial zone. This paper proposes a novel AI-driven multi-objective optimization approach for energy system design in industrial zone. Economy, environmental sustainability, and safety are considered as optimization objectives, and the Pareto front solution set is calculated using the NSGA-II algorithm. The optimal system design is determined based on the composite optimization objective function under different priority scenarios. This method provides valuable insights that could serve as a blueprint for energy system design in industrial zones, offering guidance towards more efficient and environmentally conscious energy configurations.

At present, studies of rural texture lack practical guiding significance for rural planning, resulting in a single form of new rural residential areas, causing serious damage to the original rural texture, moreover, there are many research methods for spatial planning, of which the method of machine learning has been very common in the design field, especially the ability to learn spatial features and the quantitative study of features. The method of using machine learning to influence decision-making in planning and design has not been deeply studied and applied in rural areas. The GANs (Generative Adversarial Networks) model is established in this paper by creating a four-level scale model of the boundary definition and factor-labeling paradigm of the rural space, clarifying the boundary, factor, and calculation form that affect the scale at all levels, and annotating the samples using the standard as a reference. Through the case study of a rural area in Shanghai, the labeling method was adjusted, various sample augmentation methods were continuously used to increase the sample size, the sample quality was adjusted by screening samples, and multiple pieces of training were finally generated to meet the needs of the new residential areas planning scheme in the given area. Through the establishment of area indicator and spacing indicator, the rationality of the scheme is quantitatively evaluated. The findings of this study show that machine learning approaches have a lot of potential for new settlement planning in rural areas; the GANs model is very effective in creating planning schemes, and this research method generates fresh ideas for rural area planning.

Indoor openness offers a critical reference to how users perceive an interior space. Past studies have accumulated a diverse range of impacting factors based on visual evidence and empirical data through simulated simplistic environments, yet have failed to associate such evidence with complex real-world spaces to provide evaluations. This study introduces a framework that combines computer vision and machine learning to quantitatively assess perceptions of openness in office spaces based on digitally generated visual data, feature extraction, and perception rating data. Machine learning models are trained to predict openness ratings based on the proportions of different elements. Our results demonstrate a method to integrate multiple elements for evaluating the perception of openness, and highlight that collective understanding of the concept of openness influences users’ perception.

This research aims to tackle the problem of the generative urban design of residential areas using a general-solving machine of mathematical programming. Residential areas on university campuses are taken as examples. As a type of urban design problem, the layout of residential areas on campuses is subject to multiple indicators and various boundary shapes. Quadratic Programming (QP) offers a representation of this problem, and with the assistance of cutting-edge mathematical programming solvers, the urban design problem with quadratic constraints can be automatically tackled. However, the difficulties lie in formulating complex boundaries, flexible building templates, and directional variability. To overcome these challenges, this research combines inside-model techniques of representation and outside-model modules utilizing geometric methods to enhance the main model of QP. A pipeline is provided to apply the approach in real urban design projects. The generated results validate the effectiveness of the enhanced model and the pipeline.

It is complex to design facade skin for different building, in view of the high operational energy consumption that accompanies buildings with excessively large window-to-wall ratios. For public buildings with, the energy load cost relatively large. There are a characteristic of facade which have high wall window rate to consume a lot of energy or increase Insulation cost due to the influence of interfaces. For the treatment of shading in summer, excessive overhang of the eaves often increases the load and structural cost of the roof or increases the external sunshade structure to increase structural cost. Integration of surface energy at the interface level, combined with the shading construction of light materials. In the current process of shortening the carbon cycle of buildings, study are exploring innovative exploration ways, focusing on building integrated sunshade and insulation for building with high window-wall-ratios. study consider the changes between winter and summer differentiated needs to propose optimized design solutions for reducing cooling and heating energy consumption under photo-thermal comfort conditions.set hina xian project as the case. Study show by new toughness skin design improving Performance and efficient cut of Energy cost. Solving photo-ermal objective analysis and selection problems based on ANN neural network learning prediction feedback, bench marking of environmental parameters, and parameter definition of evolutionary solvers. The results show that the solver can reach convergence at an early stage, and the validation of the chosen solution proves the effectiveness of the strategy-guided morphology. Based on learning prediction can be more accurately coupled with existing simulation trends. The total annual energy consumption of design scenario for this case skin, which is 8.4% more energy efficient than the conventional scenario and more than 5.3℃.

With the escalation of climate change, extreme weather events have become increasingly common, posing significant challenges to the architectural domain. Only focusing on the design methods and characteristics of buildings in typical climates is no longer enough to cope with future weather challenges. Therefore, the study of architectural design under extreme climates has become an emerging and important topic. However, China's research and engagement on extreme climate architecture lag behind many other countries, resulting in related architectural knowledge in this domain being scattered and fragmented, and even in the gray area of information retrieval, making it difficult for architects to use. This paper explores how to systematically categorize, organize, and present information on extreme climate architecture in a way that is easily accessible and beneficial to architects, thereby supporting well-informed design decisions. It proposes a top-down approach to constructing a knowledge graph for extreme climate architecture. By leveraging architectural programming theory, the study constructs an ontology model and employs ChatGPT for the extraction of knowledge from unstructured data. Additionally, it uses web crawlers to gather relevant information from general encyclopedias, integrating these into a triplet form. Compiled Data is then stored in Neo4j, facilitating efficient domain knowledge querying and visualization. Furthermore, this research presents a Q&A application demo named Extreme-Architecture-Graph based on the constructed knowledge graph. This application aims to transform extreme climate architectural knowledge into actionable insights, enabling architects to acquire a comprehensive understanding of design objectives under extreme climate conditions in the early stages of planning and design. In summary, this study constructs a knowledge graph for architecture in extreme environments to enhance preparedness for the unforeseen risks posed by extreme weather conditions, and explores the usability of multi-source heterogeneous data in the field of architecture in the era of artificial intelligence.

Nowadays, frequent local wars have inflicted severe damage on urban built environments, presenting substantial challenges for post-war restoration. Moreover, the scarcity of architectural imagery further exacerbates these challenges. In this context, virtual restoration techniques have shown significant advantages in speed and accuracy over traditional experience-based methods. This paper aims to explore the potential of artificial intelligence in the restoration of architectural ruins and the generation of visual predictions. Specifically, we compared the performance of pix2pix GAN and Stable Diffusion Models in architectural restoration, then further applied Stable Diffusion Models based on a modern style to the entire post-war restoration process spanning time. Notably, the optimization of its U-NET module through rule-enhanced learning and the precise mapping of image features through ControlNet improved the accuracy and coherence of restoration. Experimental findings indicate that Stable Diffusion Model surpasses traditional machine learning approaches in preserving architectural characteristics and styles, effectively addressing the issues of paired training data scarcity and minor facade feature dissipation, while astutely retaining selective elements indicative of war-induced architectural damage and aging.

Artificial Intelligence (AI) technology has been widely used in architectural design. However, many architects remain skeptical about AI-generated design schemes. This study aims to explore the reasons behind architects’ dissatisfaction with the AI-generated design. We invited professional and non-professional participants to evaluate three hospital design schemes: two generated by AI application EvoMass and one by experienced architects. Visual behavior data and oral description of participants during the evaluation were collected and analyzed. The results show significant differences in focal points and observation patterns. Professional participants paid more attention to the building footprint and the junction of buildings and city roads. These findings indicate that understanding of architectural design principles is crucial to enhancing Generative AI’s capabilities.

Researchers of ancient cities often rely on textual descriptions, limited archaeological excavations, and scattered map materials to infer urban forms due to the absence of intuitive materials. However, urban descriptions in historical texts are often composed of interconnected information. The intricate and complex relationships between these pieces of information form a three-dimensional and multidimensional spatiotemporal model of the ancient city. In the conventional research context, researchers need to manually combine these historical materials to make reasonable conjectures, but such conjectures are often limited by the researcher’s personal data and view, making it difficult to form a rapid and precise generation mechanism. This research introduces big data processing and large language model technologies to construct an effective symbolic database framework for conveniently analyzing, demonstrating, and utilizing Chinese ancient city information in historical texts. This framework allows for the rapid generation of different “representations” of ancient city at specific historical moments, facilitating scholarly assessments of historical scenarios, which promotes the integration of computational technology, historical urban studies, and architecture, and offering new tools and perspectives for related academic work.

Within the domain of AI-driven architectural design, the task of faithfully depicting specific architectural styles, especially in the design of building facades, continues to be a substantial difficulty. This work introduces a novel method that combines shape grammars, which is known for representing design rules, with stable diffusion models to tackle this problem. The specific focus of this study is on the Ionic style. The research pioneers a way to generate building facades by merging the technological capabilities of AI models like LoRA and Dream-Booth with the image-generating abilities of Stable Diffusion. This entails a demanding procedure of creating specialized datasets, training AI models using these datasets, and doing a thorough comparative analysis to assure the accuracy and visual authenticity of the designs in the Ionic style. The main contribution of this study is its illustration of how shape grammar may direct AI models to generate architectural facades that are of excellent quality and consistent in style. The revised model shows the ability to create Ionic-style facades with higher accuracy and AI that follows conventional architectural rules. This study highlights the capacity of AI to enhance the visual elements of architectural design, closing the divide between contemporary computational methods and conventional architectural sophistication.

The integration of Generative AI in architectural design offers unparalleled opportunities for design innovation, characterized by the creation of complex, nuanced semantic universes. This paper explores the convergence of AI with various design-based disciplines, aiming to reimagine the design process through a multi-layered strategy. It proposes a unique AI-native workflow for the seamless integration of diverse design concepts, with the primary goal of investigating the capabilities of Creative AI in architectural design and its potential to combine various design fields into a cohesive workflow that addresses the limitations of current generative AI models when applied in architectural design [1]. By employing a trifold methodology consisting of sequential city sections, semantic encoding, and crafting semantic universes, the project utilizes contextual 3D data from Chicago and conducts a comprehensive synthesis across nine design domains, thereby creating a rich ‘semantic universe’ that serves as the foundation for AI-driven design generation.

With the rapid development of information technology, the construction industry has gradually realized that the traditional two-dimensional design and construction methods can no longer meet the needs of complex projects. BIM, an integrated approach to design and management, is revolutionizing the construction industry through digital modeling, collaboration and information sharing. In this context, Revit has become one of the representative software in the field of BIM with its powerful functions and ease of use. Taking revit as an example, after a series of building prototype modeling experiments, ifc files exported from revit model are further transformed into graphic data structures, and the constructed BIM semantic graph is compared and analyzed, so as to understand revit from the perspective of the graph and enable it to organize and transmit information more comprehensively and reasonably, thus helping to build BIM information semantic graph. Facilitate the sharing and exchange of information.

With the increasing complexity of urban challenges, enhancing the quality of urban spaces has become a global priority. Urban Design, as a key discipline, has garnered significant attention. In China, Urban Design education is still in its infancy, whereas internationally, it has been evolving for nearly 70 years. Among global institutions, University College London (UCL) is renowned for its forward-thinking approach to design education. However, there is limited literature detailing the methodologies employed in teaching Urban Design at internationally acclaimed universities. This paper explores how UCL’s MArch Urban Design program cultivates students through a research-driven approach, drawing on the author’s firsthand experience. The study aims to provide insights for reforming Urban Design education in China by examining knowledge construction, teaching methods, and curriculum structure at UCL.

This paper provides an extended understanding of the categorical qualities of intentionality in Don Ihde’s Intentionality Matrix (1990) and in P.P. Verbeek’s subsequent inclusion of Hybrid and Composite Intentionality. By examining the interrogative posture of Ihde and Verbeek’s categories we find that there is insufficient understanding of the dynamic characteristics of these forms of experience. Through an explanation of a pedagogical simulation project undertaken with tertiary architecture and construction students, we provide a more accurate designation of the complex existential modes of attention in engaging with digital twin simulations which rely on authenticity of the modelled environment, and authenticity of the learning experience.

What intrinsic attributes constitute the essence of living space, and how are these spaces discerned and defined in the contemporary architectural context?This study explores the dynamic relationship between ambiance, spatial design, and human well-being within residential environments. Utilizing a multidisciplinary approach that integrates insights from psychology, architecture, and interior design, the project investigates the dynamic interplay between space structuring, functional organization, and decorative elements in shaping the ambiance and functionality of living rooms. This approach integrates empirical surveys with Generative Algorithms and a Hybrid Recommender System, representing a symbiosis of human intuition and machine precision, not only challenging traditional architectural practices and highlighting the potential of generative AI in design but also offering new insights into designing for user’s needs and emotional regulation.The implications of this research are significant for architects, designers, and stakeholders in the residential design process. It highlights the necessity of a replicable model for consensus on space-ambiance combinations, pointing towards generative design and machine learning as promising tools for bridging the gap between subjective perceptions and objective design goals.

This paper investigates the advantages of integrating a prototyping-based pedagogy into a design-build studio program. A specific design-build studio is analysed as a case study to discuss the opportunities associated with prototyping playing a more vital role in architectural education. The case study presents a unique integration of prototyping with digital fabrication, thereby enabling students to rapidly iterate designs and realise complex geometries not feasible with traditional methods. The study discusses the compatibility of prototyping-based pedagogy with digital fabrication and computational design in the educational setting. Through an analysis of the studio’s structure, the paper assesses the educational benefits associated with engaging in prototyping activities, whilst drawing comparisons to established prototyping theories from other fields. Finally, the study advocates for further research to evaluate the efficacy of prototyping-based pedagogy and develop a prototyping theory specific to architectural education.

This paper explores the development and optimization of artificial reefs by introducing a novel generative design method incorporating Computational Fluid Dynamics (CFD) and Optimization (BESO). Since the 1950s, efforts to create artificial reefs have been pursued to improve marine ecosystems. Our study first surveyed the existing design of artificial reefs. Addressing the limitations in existing design methods, this research employs a topological optimization strategy, focusing on optimizing space allocation for polyphony expansion within these reefs. By analyzing fluid dynamics and connecting it with an iterative optimization process, we assess the effectiveness of material exchange in these artificial structures. This is critical for the design process, considering constraints from advanced manufacturing to allow for quick production with natural-like geometries. To advance the design of artificial reefs and explore new possibilities, we introduce a novel generative design approach, where the design of the reef emerges from the interaction between CFD and BESO, through an iterative process of material removal and addition in response to the external loading condition. The artificial reef generated is compared against benchmarks of current artificial design to assess its material efficiency, structure performance, and geometrical characteristics in complex underwater conditions.

This paper presents a model for predicting visual comfort of natural lighting office, considering both the horizontal paper work and vertical VDT (Visual Display Terminal) work mode, based on field experiments and questionnaires. By using the prediction model, an intelligent prediction platform system for visual comfort is built. The model and platform are applied to investigate the prototype office space model in Harbin, China. By simulation and control variable method, the influence of various building facade form on the visual comfort of natural lighting office space is analyzed. Finally, this paper summarizes and proposes office building facade design strategies based on visual comfort, to enhance the overall visual comfort level of office space, and guide office building design.

For architectural design to actualize climate adaptation, it is essential to optimize building energy efficiency, emission reduction, and passive survivability. However, passive design strategies for building retrofit in the hot summer and cold winter (HSCW) zone are limited in current building energy simulation and optimization (BESO) studies, which have not been widely applied in architectural practice due to the lack of a unified standard. This paper aims to explore the effectiveness and optimization methods of passive design for the typical high-rise apartment retrofit in Philadelphia, considering the dynamic effects of energy consumption, thermal comfort, and future climate scenarios. In this study, the developed future weather files were used to plot the Givoni bioclimatic chart (GBC), and building datasets were constructed based on the EnergyPlus model simulation. Meanwhile, the optimal solutions are realized based on the Morris sensitivity analysis (SA) and NSGA-II method. The results indicate solar protection remains the most effective passive design strategy, especially for south-facing room units, while the cooling effect of natural ventilation by window opening will significantly decrease over time. It is expected that in the future, the thermal coefficient (TC) of the wall and window will increase the effectiveness of energy efficiency to 235% and 152% respectively. The combinations of passive parameters in various climatic scenarios for the overall high-rise apartment retrofit can reduce both heating and cooling loads by up to 50%, and improve the duration of passive survivability by over 400 h.

The study on the behavior of on-campus individuals provides valuable insights for campus management, resource allocation, and planning layout. The application of multi-source data offers more objective and in-depth opportunities for exploring behavioral phenomena. Focusing on the Wuxi campus of Southeast University, this research utilized Wi-Fi probe positioning technology combined with a physical environment sensor system to comprehensively collect 28.87 million positioning data points and 340,000 environmental data points over a period of 14 days. After cleaning redundant, missing, abnormal, drifting, and ping-pong data, both types of data underwent visual analysis, and their correlations were studied. Additionally, trajectory feature extraction was conducted using a convolutional autoencoder neural network. The study revealed the temporal distribution of pedestrian flow, the spatial distribution of stopover behavior, and the spatiotemporal characteristics of pedestrian trajectories. This provides a reliable basis for guiding crowd behavior by improving specific campus areas and the physical environment.

Precision maintenance and operation of building systems are crucial for energy efficiency and carbon reduction in building sector. Fluctuations in people flow play a significant role in building cooling load. Traditional air handling unit (AHU) operating under constant air volume tends to waste energy, as it lacks the capability to dynamically adjust to varying demand. This study introduces an intelligent digital twin AHU system considering the people flow variation, and the energy-saving potential is evaluated by a commercial building in Shanghai. The IoT technology is used to monitor the parameters of the AHU, using MobileNetSSD and NMS algorithms to extract people flow information from a camera. Through sensitivity analysis of the collected data, the key parameters for load prediction are identified, forming the basis for establishing a dynamic load prediction model using ANN. A digital model of the AHU is set up by TRNSYS, and the optimal control strategy is determined to minimize fan and pump energy consumption under different cooling loads. The digital twin system has been successfully implemented in a shopping mall in Shanghai, and 29.5% reduction of fan energy consumption can be achieved by dynamically adjusting air supply rate in response to the changing cooling loads.

The concept of the metaverse, representing a cutting-edge virtual reality paradigm, has gained significant traction as a mode of communication and data acquisition within virtual environments, applied across diverse domains. This study investigates the utilization of a metaverse-based platform in environmental behavior analysis, focusing on wayfinding behavior. We developed a metaverse-based platform for the Satellite Terminal 1 (S1) of Shanghai Pudong International Airport (PVG), offering participants an immersive and intuitive virtual environment. Leveraging the serious game concept, we designed a wayfinding task, through which 2,746 wayfinding trajectory data were collected for performance evaluation purposes. Notably, this metaverse-based platform presents a feasible and efficient approach for studying wayfinding behavior and streamlining data collection processes.

This research aims to investigate human perception of architectural spaces through EEG signal recognition and develop an innovative interactive-based creative mode. Artificial intelligence algorithms are increasingly used in architectural design, but they are currently limited to imitating and replicating the neural mechanisms of the biological brain. Architectural space extends beyond mere form and style. This paper explores the dynamic connection between architectural space and human cognition by utilizing brain-computer interface (BCI) technology to obtain a direct brainwave signal source. The focus is specifically on the relationship between architectural space and emotional response. Emotion recognition is achieved through the implementation of advanced deep-learning algorithms. Simultaneously, we utilize VR technology to immerse testers in the architectural environment, studying neural responses to various spaces. Our experiments have preliminary revealed spatial features that are linked to human emotions, as well as differences between the designer’s expectations and the audience’s experience. The AIGC model converts emotional data into corresponding geometric spatial features and generates new spatial scenes that contain emotional, temporal, and personal attributes. This approach comprehensively understands the interplay and co-creative essence between architecture and human consciousness, as well as externalizes human consciousness as a method in architectural design.

With advancements in biology and sensor technology, neuroscience, particularly electroencephalograph (EEG) technology, has seen significant developments and applications in the field of space emotion perception. EEG technology records brain electrical activity through electrodes placed on the scalp. The signals collected by these electrodes reflect the brain’s activity status and are utilized for studying cognition, diagnosing diseases, and more. EEG technology has revolutionized the measurement of human emotions within space contexts by providing preliminary insights into human space cognition. Traditional methods, such as subjective survey questionnaires and field visits, are characterized by high levels of subjectivity, instability, and delays, which significantly undermine the objectivity, accuracy, and real-time nature of emotion recording. In contrast, EEG technology offers real-time and objective feedback on changes in space elements, effectively guiding designers to enhance space quality and improve user satisfaction and well-being within these spaces.

Under the background of new urbanization, protection historical and cultural districts is crucial to the urban development. The purpose of this study is to explore the public’s visual preference for block space, providing empirical insights and technological innovation for protection and regeneration. A hybrid method of environmental behavior and evidence-based design theory is used to capture the visual behavior of participants in virtual reality through eye tracking technology, revealing perceptual experience and choice preference. Focused on representative historical and cultural districts in Changsha, our investigation delves into the nuanced interplay between design elements imbued with cultural significance such as architectural style, facade materials, color palettes, and street layouts and their capacity to captivate visual attention within these urban enclaves. The analysis highlights the pivotal role of culturally characteristic design elements in engaging visual interest in these districts. Furthermore, data comparison reveals a strong correlation between visual appeal, emotional resonance, and cultural identity. These insights provide a scientific basis for strategies to preserve and regenerate historic and cultural districts. This study quantitatively analyzes the challenges of block space design from a humanistic perspective. In the renewal of historical and cultural blocks, designers need to emphasize the display of historical and cultural elements in buildings and protect historical buildings in order to realize the effective inheritance and activation of blocks.

The research aims to develop a framework to address needs to improve design to production workflow efficiency in the context of high-density urban environments through the application of advanced digital tools. The work primarily focuses on delivering the work through augmented reality (AR) driven assembly processes.Due to its modularity, affordability, sustainable building practices and accessibility, the research presents the self-build home as a design scheme to demonstrate such technologies. In the design proposal, an urban village rooftop intervention was proposed, adopting the open-source WikiHouse Skylark 250 modules as the design scheme for the self-build home. An AR immersive space was developed, where assembly sequence guidance was provided to aid potential self-builders to understand the building assembly process and safely, accurately and efficiently conduct on-site assemblies.The outcomes of the research provide lasting implications to the advancement of AR-assisted production workflows, and participates in ongoing urban village regeneration efforts which supports community revitalization, increases housing opportunities and maintains its social fabric.

The food delivery industry has increasingly become an integral part of daily life for people in China. While delivery riders experience high work intensity, current urban maps lack information relevant to their fundamental needs, such as infrastructure related to safety, urban connectivity for efficiency, and environmental conditions for health. Therefore, this study focuses on the basic needs of delivery riders: safety, efficiency, and health. By referencing existing literature on bicycle index assessments, we identified three evaluation dimensions related to basic needs: infrastructure quality, connectivity, and environment. Utilizing street view images and OpenStreetMap data from Yuexiu District in Guangzhou, we extracted 14 indicators from 3 dimensions to create an urban road delivery riding index. The resulting urban delivery index map aims to assist riders in route planning, enhance map systems, and provide a foundation for future research.

In recent years, digital twin city platforms often encounter issues such as emphasizing the physical model’s accuracy over social cognition and specialized applications over a comprehensive data system, hindering the fulfilment of refined urban management’s real needs. Therefore, it is essential to define the characteristics of an urban management-oriented digital twin platform and construct a detailed evaluation mechanism. This study examines the framework for evaluating mapping rates, introducing three indicators: data resolution, data freshness, and data relevance. We developed a quantifiable and replicable evaluation model to assess data completeness, update timeliness, and network correlation degree. Using Shanghai’s Huamu digital twin platform as a case study, we calculated each indicator and formed a comprehensive mapping rate evaluation. This research achieves a quantitative analysis of digital twin city platforms’ development quality which was previously unmeasurable. Additionally, this study aids in advancing digital twin city platforms to facilitate the development of a “bottom-up” refined urban management approach.

Cycling, recognized as a healthy and environmentally beneficial mode of active transport, has gained widespread acceptance and become increasingly popular. Its prevalence is profoundly influenced by the built environment, highlighting an emerging need to explore the associations between built environment factors and cycling behaviour. With the advancement of artificial intelligence, an increasing number of scholars assess the perception of the built environment using street view imagery (SVI), analysing these perceptions in conjunction with survey data. However, the usage of real-world cycling data in assessing the built environment remains limited. In our study, we explore the relationship between the built environment and cycling behaviour by correlating image segmentation analysis results of SVI from the City of Sydney with real-world cycling data from Strava Metro Data (SMD). A multivariate Poisson regression model was applied for this research. Research findings indicate a positive correlation between cycling frequency and factors such as street greenness, presence of bike lane, traffic lights, and on-street parking, while cycling frequency is negatively associated with sky openness, enclosure, street curbs, and traffic sign frame. Therefore, to build a better cycling-friendly city, urban planners and designers should focus on factors that encourage cycling and positively influence cycling behaviour. Moreover, the novel and reliable approach of integrating SVI with real-world cycling data has potential for measuring eye-level built environments in future cycling-friendly city studies.

Urban spaces possess diverse visual qualities that significantly impact comfort, aesthetics, and navigation. This paper introduces a novel approach towards classifying urban spaces based on their visual characteristics through isovist analysis. An isovist is the polygon representing the visible areas from a given vantage point. The geometrical attributes of the isovist polygon enables a quantitative measure of visual qualities in the urban setting. However, the potential for classifying urban spaces based on the geometrical attributes of isovist polygons remains largely untapped. This paper presents a methodology to systematically categorise urban spaces using isovists and their geometrical attributes. By aggregating ten dimensions of geometrical attributes through a Gaussian Mixture Model (GMM) clustering analysis, this workflow produces a classifier that categorises urban spaces into 10 distinct spatial types, each possessing unique visual and spatial characteristics. This method successfully captures intrinsic spatial typologies across diverse urban contexts and can reflect the values embedded in urban design schemes. By facilitating meaningful and discussions in urban planning and design, this research contributes to a deeper and numerical understanding of the spatial and visual aspects of urban design. Further research avenues include the extension of this methodology to 3D analysis and refining tessellation algorithms for improved computational efficiency and accuracy.

From the perspective of behavior, the study combines multi-agent simulation and measurement index system for the waterfront area. Connecting with the waterfront space characteristics, the study combines spatial, behavioral and correlation measurement indexes to form the measurement dimensions of access-efficiency, stay-comfort and water-friendliness. Based on literature research and behavior simulation output characteristics, the study initially constructs an index system for measuring the quality of public space in urban waterfront, with a total of 18 measuring indexes. 6 typical waterfront sections along the Huangpu River are selected as samples and the indexes of each period are output through the behavior simulation model based on the Anylogic. Through the extraction of the mean value, maximum value and minimum value, an urban waterfront public space measurement index system suitable for the Huangpu River in Shanghai is finally formed. By measuring different samples and normalizing the results, it is found that stay-comfort is the most important index, while access-efficiency and water-friendliness are of less importance. The study proposes a comprehensive measurement index system from behavioral needs, which forms a more accurate and intelligent index for analyzing waterfront areas in three dimensions. In addition, based on user's recreational behavior, a dynamic measurement method is developed to provide a more efficient quantitative reference for the identification and optimization of weakness in urban waterfront.