Cumincad : CUMINCAD Papers : Search Results

CumInCAD is a Cumulative Index about publications in Computer Aided Architectural Design
supported by the sibling associations ACADIA, CAADRIA, eCAADe, SIGraDi, ASCAAD and CAAD futures

PDF papers
References

Hits 1 to 20 of 651

Reformat results as: short short into frame detailed detailed into frame

_id	acadia20_574
id	acadia20_574
authors	Nguyen, John; Peters, Brady
year	2020
title	Computational Fluid Dynamics in Building Design Practice
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 574-583.
doi	https://doi.org/10.52842/conf.acadia.2020.1.574
summary	This paper provides a state-of-the-art of computational fluid dynamics (CFD) in the building industry. Two methods were used to find this new knowledge: a series of interviews with leading architecture, engineering, and software professionals; and a series of tests in which CFD software was evaluated using comparable criteria. The paper reports findings in technology, workflows, projects, current unmet needs, and future directions. In buildings, airflow is fundamental for heating and cooling, as well as occupant comfort and productivity. Despite its importance, the design of airflow systems is outside the realm of much of architectural design practice; but with advances in digital tools, it is now possible for architects to integrate air flow into their building design workflows (Peters and Peters 2018). As Chen (2009) states, “In order to regulate the indoor air parameters, it is essential to have suitable tools to predict ventilation performance in buildings.” By enabling scientific data to be conveyed in a visual process that provides useful analytical information to designers (Hartog and Koutamanis 2000), computer performance simulations have opened up new territories for design “by introducing environments in which we can manipulate and observe” (Kaijima et al. 2013). Beyond comfort and productivity, in recent months it has emerged that air flow may also be a matter of life and death. With the current global pandemic of SARS-CoV-2, it is indoor environments where infections most often happen (Qian et al. 2020). To design architecture in a post-COVID-19 environment will require an in-depth understanding of how air flows through space.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_584
id	acadia20_584
authors	Brás, Catarina; Castelo-Branco, Renata; Menezes Leitao, António
year	2020
title	Parametric Model Manipulation in Virtual Reality
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 584-593.
doi	https://doi.org/10.52842/conf.acadia.2020.1.584
summary	Algorithmic design (AD) uses algorithms to describe architectural designs, producing results that are visual by nature and greatly benefit from immersive visualization. Having this in mind, several approaches have been developed that allow architects to access and change their AD programs in virtual reality (VR). However, programming in VR introduces a new level of complexity that hinders creative exploration. Solutions based in visual programming offer limited parameter manipulation and do not scale well, particularly when used in a remote collaboration environment, while those based in textual programming struggle to find adequate interaction mechanisms to efficiently modify existing programs in VR. This research proposes to ease the programming task for architects who wish to develop and experiment with collaborative textual-based AD in VR, by bringing together the user-friendly features of visual programming and the flexibility and scalability of textual programming. We introduce an interface for the most common parametric changes that automatically generates the corresponding code in the AD program, and a hybrid programming solution that allows participants in an immersive collaborative design experience to combine textual programming with this new visual alternative for the parametric manipulation of the design. The proposed workflow aims to foster remote collaborative work in architecture studios, offering professionals of different backgrounds the opportunity to parametrically interact with textual-based AD projects while immersed in them.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_74
id	acadia20_74
authors	Bucklin, Oliver; Born, Larissa; Körner, Axel; Suzuki, Seiichi; Vasey, Lauren; T. Gresser, Götz; Knippers, Jan; Menges,
year	2020
title	Embedded Sensing and Control
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 74-83.
doi	https://doi.org/10.52842/conf.acadia.2020.1.074
summary	This paper investigates an interactive and adaptive control system for kinetic architectural applications with a distributed sensing and actuation network to control modular fiber-reinforced composite components. The aim of the project was to control the actuation of a foldable lightweight structure to generate programmatic changes. A server parses input commands and geometric feedback from embedded sensors and online data to drive physical actuation and generate a digital twin for real-time monitoring. Physical components are origami-like folding plates of glass and carbon-fiber-reinforced plastic, developed in parallel research. Accelerometer data is analyzed to determine component geometry. A component controller drives actuators to maintain or move towards desired positions. Touch sensors embedded within the material allow direct control, and an online user interface provides high-level kinematic goals to the system. A hierarchical control system parses various inputs and determines actuation based on safety protocols and prioritization algorithms. Development includes hardware and software to enable modular expansion. This research demonstrates strategies for embedded networks in interactive kinematic structures and opens the door for deeper investigations such as artificial intelligence in control algorithms, material computation, as well as real-time modeling and simulation of structural systems.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	ijac202018403
id	ijac202018403
authors	Dagmar Reinhardt, Matthias Hank Haeusler, Kerry London, Lian Loke, Yingbin Feng, Eduardo De Oliveira Barata, Charlotte Firth, Kate Dunn, Nariddh Khean, Alessandra Fabbri, Dylan Wozniak-O’Connor and Rin Masuda
year	2020
title	CoBuilt 4.0: Investigating the potential of collaborative robotics for subject matter experts
source	International Journal of Architectural Computing vol. 18 - no. 4, 353–370
summary	Human-robot interactions can offer alternatives and new pathways for construction industries, industrial growth and skilled labour, particularly in a context of industry 4.0. This research investigates the potential of collaborative robots (CoBots) for the construction industry and subject matter experts; by surveying industry requirements and assessments of CoBot acceptance; by investing processes and sequences of work protocols for standard architecture robots; and by exploring motion capture and tracking systems for a collaborative framework between human and robot co-workers. The research investigates CoBots as a labour and collaborative resource for construction processes that require precision, adaptability and variability.Thus, this paper reports on a joint industry, government and academic research investigation in an Australian construction context. In section 1, we introduce background data to architecture robotics in the context of construction industries and reports on three sections. Section 2 reports on current industry applications and survey results from industry and trade feedback for the adoption of robots specifically to task complexity, perceived safety, and risk awareness. Section 3, as a result of research conducted in Section 2, introduces a pilot study for carpentry task sequences with capture of computable actions. Section 4 provides a discussion of results and preliminary findings. Section 5 concludes with an outlook on how the capture of computable actions provide the foundation to future research for capturing motion and machine learning.
keywords	Industry 4.0, collaborative robotics, on-site robotic fabrication, industry research, machine learning
series	journal
email
last changed	2021/06/03 23:29

_id	acadia20_406
id	acadia20_406
authors	Duong, Eric; Vercoe, Garrett; Baharlou, Ehsan
year	2020
title	Engelbart
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 406-415.
doi	https://doi.org/10.52842/conf.acadia.2020.1.406
summary	The internet has long been viewed as a cyberspace of free and collective information, allowing for an increase in the diversity of ideas and viewpoints available to the general public. However, critics argue that the emergence of personalization algorithms on social media and other internet platforms instead reduces information diversity by forming “filter bubbles"" of viewpoints similar to the user’s own. The adoption of these personalization algorithms is due in part to advancements in natural language processing, which allow for textual analysis at unprecedented scales. This paper aims to utilize natural language processing and architectural spatial principles to present social media from a collective viewpoint rather than a personalized one. To accomplish this, the paper introduces Engelbart, a data-driven agent-based system, where real-time Twitter conversations are visualized within a two-dimensional environment. This environment is interacted with by the artificial intelligence (AI) agent, Engelbart, which summarizes crowdsourced thoughts and feelings about current trending topics. The functionality of this web application comes from the natural language processing of thousands of tweets per minute throughout several layers of operations, including sentiment analysis and word embeddings. Presented as an understandable interface, it incorporates the values of cybernetics, cyberspace, agent-based modeling, and data ethics to show the potential for social media to become a more transparent space for collective discussion.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_594
id	acadia20_594
authors	Farahbakhsh, Mehdi; Kalantar, Negar; Rybkowski, Zofia
year	2020
title	Impact of Robotic 3D Printing Process Parameters on Bond Strength
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 594-603.
doi	https://doi.org/10.52842/conf.acadia.2020.1.594
summary	Additive manufacturing (AM), also known as 3D printing, offers advantages over traditional construction technologies, increasing material efficiency, fabrication precision, and speed. However, many AM projects in academia and industrial institutions do not comply with building codes. Consequently, they are not considered safe structures for public utilization and have languished as exhibition prototypes. While three discrete scales—micro, mezzo, and macro—are investigated for AM with paste in this paper, structural integrity has been tackled on the mezzo scale to investigate the impact of process parameters on the bond strength between layers in an AM process. Real-world material deposition in a robotic-assisted AM process is subject to environmental factors such as temperature, humidity, the load of upper layers, the pressure of the nozzle on printed layers, etc. Those factors add a secondary geometric characteristic to the printed objects that was missing in the initial digital model. This paper introduces a heuristic workflow for investigating the impacts of three selective process parameters on the bond strength between layers of paste in the robotic-assisted AM of large-scale structures. The workflow includes a method for adding the secondary geometrical characteristic to the initial 3D model by employing X-ray computerized tomography (CT) scanning, digital image processing, and 3D reconstruction. Ultimately, the proposed workflow offers a pattern library that can be used by an architect or artificial intelligence (AI) algorithms in automated AM processes to create robust architectural forms.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_416
id	acadia20_416
authors	Genadt, Ariel
year	2020
title	Discrete Continuity in the Urban Architectures of H. Hara & K. Kuma
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 416-424.
doi	https://doi.org/10.52842/conf.acadia.2020.1.416
summary	The 2020 pandemic has laid bare the ambiguous value of the virtual proximity that distributed computing enables. The remote interaction it ushered in at an unprecedented scale also spawned social isolation, which is symbolically underscored by the reliance of this form of connectivity on individuals’ discrete digital identification. This cyber-spatial dualism may be called ‘discrete continuity,’ and it already appeared in architectural thought in the 1960s with the advent of cybernetics and the first computers. The duality resurfaced in the 1990s in virtual projects, when architectural software was first widely commercialized, and it reappeared in built form in the past decade. This paper sheds light on the architectural aspects of this conceptual duality by identifying the use of discreteness and continuity in the theories of two Japanese architects, Hiroshi Hara (b.1936) and his former student, Kengo Kuma (b.1954), in their attempts to combine the two topological conditions as metaphors of societal structures. They demonstrate that the onset of the current condition, while new in its pervasiveness, has been latent in architectural thinking for several decades. This paper examines Hara’s and Kuma’s theories in light of the author’s interviews with the architects, their writings, and specific projects that illustrate metaphoric translations of topological terms into social structures, reflected in turn in the organization of urban schemes and building parts. While Hara’s and Kuma’s respective implementations are poles apart visually and materially, they share the idea that the discrete continuity of contemporary urban experience ought to be reflected in architecture. This link between their ideas has previously been overlooked.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_698
id	acadia20_698
authors	Kimm, Geoff; Burry, Mark
year	2020
title	Steering into the Skid
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 698-707.
doi	https://doi.org/10.52842/conf.acadia.2020.1.698
summary	What if any perceived risks of lost authorship and artistic control posed by a wholesale embrace of artificial intelligence by the architectural profession were instead opportunities? AI’s potential to automate design has been pursued for over 50 years, yet aspirations of early researchers are not fully realized. Nonetheless, AI’s advances continue to be rapid; it is an increasingly viable adjunct to architectural practice, and there are fundamental reasons for why the perceived “risks” of AI cannot be dismissed lightly. Architects’ professional role at the intersection of social issues and technology, however, may allow them to avoid the obsolescence faced by other roles. To do this, we propose architects responsively arbitrage an ever-changing gap between maturing AI and mutable social expectations— arbitrage in the sense of seeking to exercise individual judgment to negotiate between diverse considerations and capacities for mutual advantage. Rather than feel threatened, evolving architectural practice can augment an expanded design process to generate and embed new subtleties and expectations that society may judge contemporary AI alone as being unable to achieve. Although there can be no road map to the future of AI in architecture, historical misevaluations of machines and our own human capabilities inhibit the intertwined, synergistic, and symbiotic union with AI needed to avoid a zero-sum confrontation. To act myopically, defensively, or not at all risks straitjacketing future definitions of what it means to be an architect, designer, or even a professionally unaligned creative and productive human being.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_668
id	acadia20_668
authors	Pasquero, Claudia; Poletto, Marco
year	2020
title	Deep Green
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 668-677.
doi	https://doi.org/10.52842/conf.acadia.2020.1.668
summary	Ubiquitous computing enables us to decipher the biosphere’s anthropogenic dimension, what we call the Urbansphere (Pasquero and Poletto 2020). This machinic perspective unveils a new postanthropocentric reality, where the impact of artificial systems on the natural biosphere is indeed global, but their agency is no longer entirely human. This paper explores a protocol to design the Urbansphere, or what we may call the urbanization of the nonhuman, titled DeepGreen. With the development of DeepGreen, we are testing the potential to bring the interdependence of digital and biological intelligence to the core of architectural and urban design research. This is achieved by developing a new biocomputational design workflow that enables the pairing of what is algorithmically drawn with what is biologically grown (Pasquero and Poletto 2016). In other words, and more in detail, the paper will illustrate how generative adversarial network (GAN) algorithms (Radford, Metz, and Soumith 2015) can be trained to “behave” like a Physarum polycephalum, a unicellular organism endowed with surprising computational abilities and self-organizing behaviors that have made it popular among scientist and engineers alike (Adamatzky 2010) (Fig. 1). The trained GAN_Physarum is deployed as an urban design technique to test the potential of polycephalum intelligence in solving problems of urban remetabolization and in computing scenarios of urban morphogenesis within a nonhuman conceptual framework.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_238
id	acadia20_238
authors	Zhang, Hang
year	2020
title	Text-to-Form
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 238-247.
doi	https://doi.org/10.52842/conf.acadia.2020.1.238
summary	Traditionally, architects express their thoughts on the design of 3D architectural forms via perspective renderings and standardized 2D drawings. However, as architectural design is always multidimensional and intricate, it is difficult to make others understand the design intention, concrete form, and even spatial layout through simple language descriptions. Benefiting from the fast development of machine learning, especially natural language processing and convolutional neural networks, this paper proposes a Linguistics-based Architectural Form Generative Model (LAFGM) that could be trained to make 3D architectural form predictions based simply on language input. Several related works exist that focus on learning text-to-image generation, while others have taken a further step by generating simple shapes from the descriptions. However, the text parsing and output of these works still remain either at the 2D stage or confined to a single geometry. On the basis of these works, this paper used both Stanford Scene Graph Parser (Sebastian et al. 2015) and graph convolutional networks (Kipf and Welling 2016) to compile the analytic semantic structure for the input texts, then generated the 3D architectural form expressed by the language descriptions, which is also aided by several optimization algorithms. To a certain extent, the training results approached the 3D form intended in the textual description, not only indicating the tremendous potential of LAFGM from linguistic input to 3D architectural form, but also innovating design expression and communication regarding 3D spatial information.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	ijac202018402
id	ijac202018402
authors	Mette Ramsgaard Thomsen, Paul Nicholas, Martin Tamke, Sebastian Gatz, Yuliya Sinke and Gabriella Rossi
year	2020
title	Towards machine learning for architectural fabrication in the age of industry 4.0
source	International Journal of Architectural Computing vol. 18 - no. 4, 335–352
summary	Machine Learning (ML) is opening new perspectives for architectural fabrication, as it holds the potential for the profession to shortcut the currently tedious and costly setup of digital integrated design to fabrication workflows and make these more adaptable. To establish and alter these workflows rapidly becomes a main concern with the advent of Industry 4.0 in building industry. In this article we present two projects, which presents how ML can lead to radical changes in generation of fabrication data and linking these directly to design intent. We investigate two different moments of implementation: linking performance to the generation of fabrication data (KnitCone) and integrating the ability to adapt fabrication data in realtime as response to fabrication processes (Neural-Network Steered Robotic Fabrication). Together they examine how models can employ design information as training data and be trained to by step processes within the digital chain. We detail the advantages and limitations of each experiment, we reflect on core questions and perspectives of ML for architectural fabrication: the nature of data to be used, the capacity of these algorithms to encode complexity and generalize results, their task-specificness versus their adaptability and the tradeoffs of using them with respect to conventional explicit analytical modelling.
keywords	Machine learning, architectural design, industry 4.0, digital fabrication, robotic fabrication, CNC knit, neural networks
series	journal
email
last changed	2021/06/03 23:29

_id	acadia20_516
id	acadia20_516
authors	Aghaei Meibodi, Mania; Voltl, Christopher; Craney, Ryan
year	2020
title	Additive Thermoplastic Formwork for Freeform Concrete Columns
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 516-525.
doi	https://doi.org/10.52842/conf.acadia.2020.1.516
summary	The degree of geometric complexity a concrete element can assume is directly linked to our ability to fabricate its formwork. Additive manufacturing allows fabrication of freeform formwork and expands the design possibilities for concrete elements. In particular, fused deposition modeling (FDM) 3D printing of thermoplastic is a useful method of formwork fabrication due to the lightweight properties of the resulting formwork and the accessibility of FDM 3D printing technology. The research in this area is in early stages of development, including several existing efforts examining the 3D printing of a single material for formwork— including two medium-scale projects using PLA and PVA. However, the performance of 3D printed formwork and its geometric complexity varies, depending on the material used for 3D printing the formwork. To expand the existing research, this paper reviews the opportunities and challenges of using 3D printed thermoplastic formwork for fabricating custom concrete elements using multiple thermoplastic materials. This research cross-references and investigates PLA, PVA, PETG, and the combination of PLA-PVA as formwork material, through the design and fabrication of nonstandard structural concrete columns. The formwork was produced using robotic pellet extrusion and filament-based 3D printing. A series of case studies showcase the increased geometric freedom achievable in formwork when 3D printing with multiple materials. They investigate the potential variations in fabrication methods and their print characteristics when using different 3D printing technologies and printing materials. Additionally, the research compares speed, cost, geometric freedom, and surface resolution.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_456
id	acadia20_456
authors	Alali, Jiries; Negar Kalantar, Dr.; Borhani, Alireza
year	2020
title	Casting on a Dump
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 456-463.
doi	https://doi.org/10.52842/conf.acadia.2020.1.456
summary	“Casting on a dump” focuses on finding accessible, low-tech fabrication methodologies that allow for the construction of parametrically designed nonstandard modular cast panels. Such an approach adopts a computational design framework using a single low-tech and low-energy fabrication device to create nonrepetitive volumetric panels cast in situ. The design input for these panels is derived from design preferences and environmental control data. The technique expands upon easy to fabricate and cast methods, targeting less-developed logistical settings worldwide, and thus responding to imminent needs related to climate, available resources, and the economy.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_228
id	acadia20_228
authors	Alawadhi, Mohammad; Yan, Wei
year	2020
title	BIM Hyperreality
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 228-236.
doi	https://doi.org/10.52842/conf.acadia.2020.1.228
summary	Deep learning is expected to offer new opportunities and a new paradigm for the field of architecture. One such opportunity is teaching neural networks to visually understand architectural elements from the built environment. However, the availability of large training datasets is one of the biggest limitations of neural networks. Also, the vast majority of training data for visual recognition tasks is annotated by humans. In order to resolve this bottleneck, we present a concept of a hybrid system—using both building information modeling (BIM) and hyperrealistic (photorealistic) rendering—to synthesize datasets for training a neural network for building object recognition in photos. For generating our training dataset, BIMrAI, we used an existing BIM model and a corresponding photorealistically rendered model of the same building. We created methods for using renderings to train a deep learning model, trained a generative adversarial network (GAN) model using these methods, and tested the output model on real-world photos. For the specific case study presented in this paper, our results show that a neural network trained with synthetic data (i.e., photorealistic renderings and BIM-based semantic labels) can be used to identify building objects from photos without using photos in the training data. Future work can enhance the presented methods using available BIM models and renderings for more generalized mapping and description of photographed built environments.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	ecaade2020_193
id	ecaade2020_193
authors	Alymani, Abdulrahman, Jabi, Wassim and Corcoran, Padraig
year	2020
title	Machine Learning Methods for Clustering Architectural Precedents - Classifying the relationship between building and ground
source	Werner, L and Koering, D (eds.), Anthropologic: Architecture and Fabrication in the cognitive age - Proceedings of the 38th eCAADe Conference - Volume 1, TU Berlin, Berlin, Germany, 16-18 September 2020, pp. 643-652
doi	https://doi.org/10.52842/conf.ecaade.2020.1.643
summary	Every time an object is built, it creates a relationship with the ground. Architects have a full responsibility to design the building by taking the ground into consideration. In the field of architecture, using data mining to identify any unusual patterns or emergent architectural trends is a nascent area that has yet to be fully explored. Clustering techniques are an essential tool in this process for organising large datasets. In this paper, we propose a novel proof-of-concept workflow that enables a machine learning computer system to cluster aspects of an architect's building design style with respect to how the buildings in question relate to the ground. The experimental workflow in this paper consists of two stages. In the first stage, we use a database system to collect, organise and store several significant architectural precedents. The second stage examines the most well-known unsupervised learning algorithm clustering techniques which are: K-Means, K-Modes and Gaussian Mixture Models. Our experiments demonstrated that the K-means clustering algorithm method achieves a level of accuracy that is higher than other clustering methods. This research points to the potential of AI in helping designers identify the typological and topological characteristics of architectural solutions and place them within the most relevant architectural canons
keywords	Machine Learning; Building and Ground Relationship; Clustering Algorithms; K-means cluster Algorithms
series	eCAADe
email
last changed	2022/06/07 07:54

_id	acadia20_350
id	acadia20_350
authors	Atanasova, Lidia; Mitterberger, Daniela; Sandy, Timothy; Gramazio, Fabio; Kohler, Matthias; Dörfler, Kathrin
year	2020
title	Prototype As Artefact
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 350-359.
doi	https://doi.org/10.52842/conf.acadia.2020.1.350
summary	In digital design-to-fabrication workflows in architecture, in which digitally controlled machines perform complex fabrication tasks, all design decisions are typically made before production. In such processes, the formal definition of the final shape is explicitly inscribed into the design model by means of corresponding step-by-step machine instructions. The increasing use of augmented reality (AR) technologies for digital fabrication workflows, in which people are instructed to carry out complex fabrication tasks via AR interfaces, creates an opportunity to question and adjust the level of detail and the nature of such explicit formal definitions. People’s cognitive abilities could be leveraged to integrate explicit machine intelligence with implicit human knowledge and creativity, and thus to open up digital fabrication to intuitive and spontaneous design decisions during the building process. To address this question, this paper introduces open-ended Prototype-as-Artefact fabrication workflows that examine the possibilities of designing and creative choices while building in a human-robot collaborative setting. It describes the collaborative assembly of a complex timber structure with alternating building actions by two people and a collaborative robot, interfacing via a mobile device with object tracking and AR visualization functions. The spatial timber assembly being constructed follows a predefined grammar but is not planned at the beginning of the process; it is instead designed during fabrication. Prototype-as-Artefact thus serves as a case study to probe the potential of both intuitive and rational aspects of building and to create new collaborative work processes between humans and machines.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_66
id	acadia20_66
authors	Aviv, Dorit; Wang, Zherui; Meggers, Forrest; Ida, Aletheia
year	2020
title	Surface Generation of Radiatively-Cooled Building Skin for Desert Climate
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 66-73.
doi	https://doi.org/10.52842/conf.acadia.2020.1.066
summary	A radiatively cooled translucent building skin is developed for desert climates, constructed out of pockets of high heat-capacity liquids. The liquids are contained by a wavelength-selective membrane enclosure, which is transmissive in the infrared range of electromagnetic radiation but reflective in the shortwave range, and therefore prevents overheating from solar radiation and at the same time allows for passive cooling through exposure of its thermal mass to the desert sky. To assess the relationship between the form and performance of this envelope design, we develop a feedback loop between computational simulations, analytical models, and physical tests. We conduct a series of simulations and bench-scale experiments to determine the thermal behavior of the proposed skin and its cooling potential. Several materials are considered for their thermal storage capacity. Hydrogel cast into membrane enclosures is tested in real climate conditions. Slurry phase change materials (PCM) are also considered for their additional heat storage capacity. Challenges of membrane welding patterns and nonuniform expansion of the membrane due to the weight of the enclosed liquid are examined in both digital simulations and physical experiments. A workflow is proposed between the radiation analysis based on climate data, the formfinding simulations of the elastic membrane under the liquid weight, and the thermal storage capacity of the overall skin.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_202p
id	acadia20_202p
authors	Battaglia, Christopher A.; Verian, Kho; Miller, Martin F.
year	2020
title	DE:Stress Pavilion
source	ACADIA 2020: Distributed Proximities / Volume II: Projects [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95253-6]. Online and Global. 24-30 October 2020. edited by M. Yablonina, A. Marcus, S. Doyle, M. del Campo, V. Ago, B. Slocum. 202-207
summary	Print-Cast Concrete investigates concrete 3D printing utilizing robotically fabricated recyclable green sand molds for the fabrication of thin shell architecture. The presented process expedites the production of doubly curved concrete geometries by replacing traditional formwork casting or horizontal corbeling with spatial concrete arching by developing a three-dimensional extrusion path for deposition. Creating robust non-zero Gaussian curvature in concrete, this method increases fabrication speed for mass customized elements eliminating two-part mold casting by combining robotic 3D printing and extrusion casting. Through the casting component of this method, concrete 3D prints have greater resolution along the edge condition resulting in tighter assembly tolerances between multiple aggregated components. Print-Cast Concrete was developed to produce a full-scale architectural installation commissioned for Exhibit Columbus 2019. The concrete 3D printed compression shell spanned 12 meters in length, 5 meters in width, and 3 meters in height and consisted of 110 bespoke panels ranging in weight of 45 kg to 160 kg per panel. Geometrical constraints were determined by the bounding box of compressed sand mold blanks and tooling parameters of both CNC milling and concrete extrusion. Using this construction method, the project was able to be assembled and disassembled within the timeframe of the temporary outdoor exhibit, produce <1% of waste mortar material in fabrication, and utilize 60% less material to construct than cast-in-place construction. Using the sand mold to contain geometric edge conditions, the Print-Cast technique allows for precise aggregation tolerances. To increase the pavilions resistance to shear forces, interlocking nesting geometries are integrated into each edge condition of the panels with .785 radians of the undercut. Over extruding strategically during the printing process casts the undulating surface with accuracy. When nested together, the edge condition informs both the construction logic of the panel’s placement and orientation for the concrete panelized shell.
series	ACADIA
type	project
email
last changed	2021/10/26 08:08

_id	acadia20_526
id	acadia20_526
authors	Bruce, Mackenzie; Clune, Gabrielle; Culligan, Ryan; Vansice, Kyle; Attraya, Rahul; McGee, Wes; Yan Ng, Tsz
year	2020
title	FORM{less}
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 526-535
doi	https://doi.org/10.52842/conf.acadia.2020.1.526
summary	Form{less} focuses on the creation of complex thin-shell concrete forms using robotically thermoformed plastic molds. Typically, similar molds would be created using the vacuum forming process, producing direct replications of the pattern. Creating molds with this process is not only time- and material-intensive but also costly if customization is involved. Thin-shell concrete forms often require a labor-intensive process of manually finishing the open-face surface. The devised process of thermoforming two nested molds allows the concrete to be cast in between, with finished surfaces on both sides. Molds made with polyethylene terephthalate glycol (PETG) allow the formwork to be reused and recycled. The research and fabrication work include the development of heating elements and the creation of the robotic process for forming the PETG. The PETG is manipulated via a robotic arm, with a custom magnetic end effector. The integration of robotics not only enables precision for manufacturing but also allows for replicability with unrestricted threedimensional deformation. The repeatable process allows for rapid prototyping and geometric customization. Design options are then simulated computationally using SuperMatterTools, enabling further design exploration of this process without the need for extensive physical prototyping. This research aims to develop a process that allows for the creation of complex geometries while reducing the amount of material waste used for concrete casting. The novelty of the process created by dynamically forming PETG allows for quick production of formwork that is both customizable and replicable. This method of creating double-sided building components is simulated at various scales of implementation.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	caadria2020_224
id	caadria2020_224
authors	Castelo-Branco, Renata and LeitÃ£o, AntÃ³nio
year	2020
title	Visual Meets Textual - A Hybrid Programming Environment for Algorithmic Design
source	D. Holzer, W. Nakapan, A. Globa, I. Koh (eds.), RE: Anthropocene, Design in the Age of Humans - Proceedings of the 25th CAADRIA Conference - Volume 1, Chulalongkorn University, Bangkok, Thailand, 5-6 August 2020, pp. 375-384
doi	https://doi.org/10.52842/conf.caadria.2020.1.375
summary	Algorithmic approaches are currently being introduced in many areas of human activity and architecture is no exception. However, designing with algorithms is a foreign concept to many and the inadequacy of current programming environments creates a barrier to the generalized adoption of Algorithmic Design (AD). This research aims to provide architects with a programming tool they feel comfortable with, while allowing them to fully benefit from AD's advantages in the creation of complex architectural models. We present Khepri.gh, a hybrid solution that combines Grasshopper, a visual programming environment, with Khepri, a flexible and scalable textual programming tool. Khepri.gh establishes a bridge between the visual and the textual paradigm, offering its users the best of both worlds while providing an extra set of advantages, including portability among CAD, BIM, and analysis tools.
keywords	Algorithmic Design; Hybrid Programming Environment; Textual Programming; Visual Programming
series	CAADRIA
email
last changed	2022/06/07 07:55

For more results click below: