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 370

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

_id	ecaade2021_148
id	ecaade2021_148
authors	Mintrone, Alessandro and Erioli, Alessio
year	2021
title	Training Spaces - Fostering machine sensibility for spatial assemblages through wave function collapse and reinforcement learning
source	Stojakovic, V and Tepavcevic, B (eds.), Towards a new, configurable architecture - Proceedings of the 39th eCAADe Conference - Volume 1, University of Novi Sad, Novi Sad, Serbia, 8-10 September 2021, pp. 17-26
doi	https://doi.org/10.52842/conf.ecaade.2021.1.017
summary	This research explores the integration of Deep Reinforcement Learning (RL) and a Wave Function Collapse (WFC) algorithm for a goal-driven, open-ended generation of architectural spaces. Our approach binds RL to a distributed network of decisions, unfolding through three key steps: the definition of a set of architectural components (tiles) and their connectivity rules, the selection of the tile placement location, which is determined by the WFC, and the choice of which tile to place, which is performed by RL. The act of thinking becomes granular and embedded in an iterative process, distributed among human and non-human cognitions, which constantly negotiate their agency and authorial status. Tools become active agents capable of developing their own sensibility while controlling specific spatial conditions. Establishing an interdependency with the human, that engenders the design patterns and becomes an indispensable prerequisite for the exploration of the generated design space, exceeding human or machinic reach alone.
keywords	Reinforcement Learning; Machine Learning; Proximal Policy Optimization; Assemblages; Wave Function Collapse
series	eCAADe
email
last changed	2022/06/07 07:58

_id	ecaade2021_247
id	ecaade2021_247
authors	Wibranek, Bastian, Liu, Yuxi, Funk, Niklas, Belousov, Boris, Peters, Jan and Tessmann, Oliver
year	2021
title	Reinforcement Learning for Sequential Assembly of SL-Blocks - Self-interlocking combinatorial design based on Machine Learning
source	Stojakovic, V and Tepavcevic, B (eds.), Towards a new, configurable architecture - Proceedings of the 39th eCAADe Conference - Volume 1, University of Novi Sad, Novi Sad, Serbia, 8-10 September 2021, pp. 27-36
doi	https://doi.org/10.52842/conf.ecaade.2021.1.027
summary	Adaptive reconfigurable structures are seen as the next big step in the evolution of architecture. However, to achieve this vision, new tools are required that enable autonomous configuration of given elements based on a specified design objective. Various approaches have been considered in the past, ranging from rule-based methods to evolutionary optimization. Although successful in applications where search heuristics or informative objective functions can be provided, these methods struggle with long-term planning problems. In this paper, we tackle the problem of sequential assembly of SL-blocks which has the character of a combinatorial optimization problem. We explore the applicability of deep reinforcement learning algorithms that recently showed great success on combinatorial problems in other domains, such as board games and molecular design. We highlight the unique challenges presented by the architectural design setting and compare the performance to evolutionary computation and heuristic search baselines.
keywords	Reinforcement Learning; Architectural Assembly; Discrete Design; SL-blocks; Dry Joined
series	eCAADe
email
last changed	2022/06/07 07:57

_id	ecaade2021_150
id	ecaade2021_150
authors	Song, Yanan and Yuan, Philip F.
year	2021
title	A Research On Building Cluster Morphology Formation Based On Wind Environmental Performance And Deep Reinforcement Learning
source	Stojakovic, V and Tepavcevic, B (eds.), Towards a new, configurable architecture - Proceedings of the 39th eCAADe Conference - Volume 1, University of Novi Sad, Novi Sad, Serbia, 8-10 September 2021, pp. 335-344
doi	https://doi.org/10.52842/conf.ecaade.2021.1.335
summary	Nowadays, numerous researchers emphasize the significance of the environmen-tal performance-driven generative methodology. However, due to the complex coupling mechanism of environmental regulation factors, the existing optimiza-tion engines and applications are time-consuming and cumbersome. In this re-search, we propose a novel design methodology based on Deep Reinforcement Learning (DRL). This paper is divided into 3 sections, including theoretical framework, design strategy, and practical application. It first introduces an over-view of basic principles, illustrating the potential advantages of DRL in perfor-mance data-driven design. Based on this, the paper proposes a DRL-based gener-ative method. We point out a more specific discussion about the application and workflow of core DRL elements in architectural design. Finally, taking a grid-form urban space composed by multitude high-rise building blocks as an exam-ple, we present a application through a DRL agent to conduct numerous active wind environmental performance-based design tests. It is an interactive and gen-erative design method, owning multiple advantages of timeliness, convenience, and intelligence.
keywords	Deep Reinforcement Learning; Environmental Performance Design; Generative Design; Building Cluster Formation
series	eCAADe
email
last changed	2022/06/07 07:56

_id	acadia21_470
id	acadia21_470
authors	£ochnicki, Grzegorz; Kalousdian, Nicolas Kubail; Leder, Samuel; Maierhofer, Mathias; Wood, Dylan; Menges, Achim
year	2021
title	Co-Designing Material-Robot Construction Behaviors
source	ACADIA 2021: Realignments: Toward Critical Computation [Proceedings of the 41st Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 979-8-986-08056-7]. Online and Global. 3-6 November 2021. edited by B. Bogosian, K. Dörfler, B. Farahi, J. Garcia del Castillo y López, J. Grant, V. Noel, S. Parascho, and J. Scott. 470-479.
doi	https://doi.org/10.52842/conf.acadia.2021.470
summary	This paper presents research on designing distributed, robotic construction systems in which robots are taught construction behaviors relative to the elastic bending of natural building materials. Using this behavioral relationship as a driver, the robotic system is developed to deal with the unpredictability of natural materials in construction and further to engage their dynamic characteristics as methods of locomotion and manipulation during the assembly of actively bent structures. Such an approach has the potential to unlock robotic building practice with rapid-renewable materials, whose short crop cycles and small carbon footprints make them particularly important inroads to sustainable construction. The research is conducted through an initial case study in which a mobile robot learns a control policy for elastically bending bamboo bundles into designed configurations using deep reinforcement learning algorithms. This policy is utilized in the process of designing relevant structures, and for the in-situ assembly of these designs. These concepts are further investigated through the co-design and physical prototyping of a mobile robot and the construction of bundled bamboo structures. This research demonstrates a shift from an approach of absolute control and predictability to behavior-based methods of assembly. With this, materials and processes that are often considered too labor-intensive or unpredictable can be reintroduced. This reintroduction leads to new insights in architectural design and construction, where design outcome is uniquely tied to the building material and its assembly logic. This highly material-driven approach sets the stage for developing an effective, sustainable, light-touch method of building using natural materials.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	cdrf2021_26
id	cdrf2021_26
authors	Yuqian Li and Weiguo Xu
year	2021
title	Using CycleGAN to Achieve the Sketch Recognition Process of Sketch-Based Modeling
source	Proceedings of the 2021 DigitalFUTURES The 3rd International Conference on Computational Design and Robotic Fabrication (CDRF 2021)
doi	https://doi.org/https://doi.org/10.1007/978-981-16-5983-6_3
summary	Architects usually design ideation and conception by hand-sketching. Sketching is a direct expression of the architect’s creativity. But 2D sketches are often vague, intentional and even ambiguous. In the research of sketch-based modeling, it is the most difficult part to make the computer to recognize the sketches. Because of the development of artificial intelligence, especially deep learning technology, Convolutional Neural Networks (CNNs) have shown obvious advantages in the field of extracting features and matching, and Generative Adversarial Neural Networks (GANs) have made great breakthroughs in the field of architectural generation which make the image-to-image translation become more and more popular. As the building images are gradually developed from the original sketches, in this research, we try to develop a system from the sketches to the images of buildings using CycleGAN algorithm. The experiment demonstrates that this method could achieve the mapping process from the sketches to images, and the results show that the sketches’ features could be recognised in the process.By the learning and training process of the sketches’ reconstruction, the features of the images are also mapped to the sketches, which strengthen the architectural relationship in the sketch, so that the original sketch can gradually approach the building images, and then it is possible to achieve the sketch-based modeling technology.
series	cdrf
email
last changed	2022/09/29 07:53

_id	ascaad2021_151
id	ascaad2021_151
authors	Allam, Samar; Soha El Gohary, Maha El Gohary
year	2021
title	Surface Shape Grammar Morphology to Optimize Daylighting in Mixed-Use Building Skin
source	Abdelmohsen, S, El-Khouly, T, Mallasi, Z and Bennadji, A (eds.), Architecture in the Age of Disruptive Technologies: Transformations and Challenges [9th ASCAAD Conference Proceedings ISBN 978-1-907349-20-1] Cairo (Egypt) [Virtual Conference] 2-4 March 2021, pp. 479-492
summary	Building Performance simulation is escalating towards design optimization worldwide utilizing computational and advanced tools. Egypt has its plan and agenda to adopt new technologies to mitigate energy consumption through various sectors. Energy consumption includes electricity, crude oil, it encompasses renewable and non-renewable energy consumption. Egypt Electricity (EE) consumption by sector percentages is residential (47%), industrial (25%) and commercial (12%), with the remainder used by government, agriculture, public lighting and public utilities (4%). Electricity building consumption has many divisions includes HVAC systems, lighting, Computers and Electronics and others. Lighting share of electricity consumption can vary from 11 to 15 percent in mixed buildings as in our case study which definitely less that the amount used for HVAC loads. This research aims at utilizing shape morphogenesis on facades using geometric shape grammar to enhance daylighting while blocking longwave radiations causing heat stress. Mixed-use building operates in daytime more than night which emphasizes the objective of this study. Results evaluation is referenced to LEED v4.1 and ASHRAE 90.1-2016 window-to-wall ratio calibration and massive wall description. Geometric morphogenesis relies on three main parameters; Pattern (Geometry Shape Grammar: R1, R2, and R3), a reference surface to map from, and a target surface to map to which is the south-western façade of the case study. Enhancing Geo-morph rule is to guarantee flexibility due to the rotation of sun path annually with different azimuth and altitude angles and follow LEED V4.1 enhancements of opaque wall percent for building envelope.
series	ASCAAD
email
last changed	2021/08/09 13:13

_id	caadria2021_136
id	caadria2021_136
authors	Carallo, Marinella
year	2021
title	Office building design in Hong Kong Island through shape optimization
source	A. Globa, J. van Ameijde, A. Fingrut, N. Kim, T.T.S. Lo (eds.), PROJECTIONS - Proceedings of the 26th CAADRIA Conference - Volume 1, The Chinese University of Hong Kong and Online, Hong Kong, 29 March - 1 April 2021, pp. 441-450
doi	https://doi.org/10.52842/conf.caadria.2021.1.441
summary	Dealing with crucial decision-making process has led to the development of many different methods of multicriteria assessments, especially optimization methodologies. This work is mainly focused on the integration of advanced computational design and digital methods, to design a complex building shape resulting in a performance-based approach through optimization methodologies. The project consists of the design of a skyscraper in Hong Kong Island made through parametrically controlled shape and evaluated respect to light and wind to reduce Urban Heat Island phenomena and enhance liveability. The aim is to find out a unique methodology that can be applied to different cases by making small adaptations regarding the parametrization and the parameters involved. The design is divided into two stages that need to arrange the methodology at different levels throughout the workflow. For this reason, it is mandatory to adapt inputs to the algorithm according to the goal. The result is a skyscraper placed in the financial district of Hong Kong, which has both the features of a Grade A Office building and can mitigate the UHI effect thanks to its particular and optimized shape.
keywords	shape optimization; Computational design; Genetic Algorithm; UHI effect; ventilation
series	CAADRIA
email
last changed	2022/06/07 07:54

_id	caadria2021_389
id	caadria2021_389
authors	del Campo, Matias
year	2021
title	Architecture,Language and AI - Language,Attentional Generative Adversarial Networks (AttnGAN) and Architecture Design
source	A. Globa, J. van Ameijde, A. Fingrut, N. Kim, T.T.S. Lo (eds.), PROJECTIONS - Proceedings of the 26th CAADRIA Conference - Volume 1, The Chinese University of Hong Kong and Online, Hong Kong, 29 March - 1 April 2021, pp. 211-220
doi	https://doi.org/10.52842/conf.caadria.2021.1.211
summary	The motivation to explore Attentional Generative Adversarial Networks (AttnGAN) as a design technique in architecture can be found in the desire to interrogate an alternative design methodology that does not rely on images as starting point for architecture design, but language. Traditionally architecture design relies on visual language to initiate a design process, wither this be a napkin sketch or a quick doodle in a 3D modeling environment. AttnGAN explores the information space present in programmatic needs, expressed in written form, and transforms them into a visual output. The key results of this research are shown in this paper with a proof-of-concept project: the competition entry for the 24 Highschool in Shenzhen, China. This award-winning project demonstrated the ability of GraphCNN to serve as a successful design methodology for a complex architecture program. In the area of Neural Architecture, this technique allows to interrogate shape through language. An alternative design method that creates its own unique sensibility.
keywords	Artificial Intelligence; Machine Learning; Artificial Neural Networks; Semiotics; Design Methodology
series	CAADRIA
email
last changed	2022/06/07 07:55

_id	ascaad2021_051
id	ascaad2021_051
authors	Marey, Ahmed; Ahmed Barakat
year	2021
title	The Customized Habitat: An Exploration of Personality-Induced Mass Customization through Shape Grammars
source	Abdelmohsen, S, El-Khouly, T, Mallasi, Z and Bennadji, A (eds.), Architecture in the Age of Disruptive Technologies: Transformations and Challenges [9th ASCAAD Conference Proceedings ISBN 978-1-907349-20-1] Cairo (Egypt) [Virtual Conference] 2-4 March 2021, pp. 450-464
summary	Despite its economic efficiency, mass production fails to appeal to the very people it is meant to accommodate. Mass customization, on the other hand, allows for the consideration of personal differences. Nonetheless, it is a process that requires more time, effort, and resources, hence the reliance upon mass production. Previous research showed a potential impact of personality on perceptions of the architectural space. The research investigates the applicability of mass customization in the architectural domain using MBTI (Myers–Briggs Type Indicator). Using MBTI, we surveyed 187 individuals to investigate the correlations between personal traits (mind, energy, nature, tactics, and identity) and preferences of architectural aspects (exposure, circulation, view, plan layout, and interior colors). The survey draws on how multiple fields have successfully applied MBTI to increase the value they provide. The findings present a novel contribution to architectural research as they demonstrate an actual connection between MBTI personality patterns and architectural preferences. In addition to several interaction patterns, our results strongly support an effect of the mind aspect on view preferences as well as an effect of energy on three architectural aspects: view, plan layout and interior colors. Shape grammars were then created, based upon these correlations, in order to provide a basis for optimized mass customization. The optimization/automation of this process will result in a more habitable space in which neither personality differences nor valuable resources are sacrificed.
series	ASCAAD
email
last changed	2021/08/09 13:11

_id	ascaad2021_153
id	ascaad2021_153
authors	Valitabar, Mahdi; Mohammadjavad Mahdavinejad, Henry Skates, Peiman Pilechiha
year	2021
title	Data-Driven Design of Adaptive Façades: View, Glare, Daylighting and Energy Efficiency
source	Abdelmohsen, S, El-Khouly, T, Mallasi, Z and Bennadji, A (eds.), Architecture in the Age of Disruptive Technologies: Transformations and Challenges [9th ASCAAD Conference Proceedings ISBN 978-1-907349-20-1] Cairo (Egypt) [Virtual Conference] 2-4 March 2021, pp. 699-711
summary	This paper attempts to increase occupants’ view to outside through Adaptive facades by employing a parametric design method. Reaching a balance between occupants’ requirements and the building energy criteria is the main objective of this research. To this end, a multi-objective optimization is done to generate some optimum models. The method, indeed, was used to optimize the shading size of a dynamic vertical shading system utilized on the south façade of a single office room located in Tehran. The shading system was defined by five parameters and a combination of Cut-off and a glare protection strategy is used to control dynamic shadings. The size-optimisation objectives are minimum DGP, cooling load and maximum illuminance, which were analysed by Ladybug Tools. Then, Octopus was used for multi-objective optimistion to find new optimum forms. Along with the openness factor, a new index is presented to evaluate the outside view in multiple louver shading systems, named “Openness Curvature Factor” (OCF). Thanks to this method, the size and shape of some optimum generated models were modified to increase the amount of OCF. Following that the Honeybee Plus is used to simulate the visual performance of modified models which shows a significant improvement. The modified models could provide about 4 times more outside view than generated models whilst keeping the DGP value in imperceptible range. Geometric or even complex non-geometric shading forms can be studied by this method to find optimum adaptive facades.
series	ASCAAD
email
last changed	2021/08/09 13:14

_id	caadria2021_161
id	caadria2021_161
authors	Zhao, Xin, Han, Yunsong and Shen, Linhai
year	2021
title	Multi-objective Optimisation of a Free-form Building Shape to improve the Solar Energy Utilisation Potential using Artificial Neural Networks
source	A. Globa, J. van Ameijde, A. Fingrut, N. Kim, T.T.S. Lo (eds.), PROJECTIONS - Proceedings of the 26th CAADRIA Conference - Volume 1, The Chinese University of Hong Kong and Online, Hong Kong, 29 March - 1 April 2021, pp. 221-230
doi	https://doi.org/10.52842/conf.caadria.2021.1.221
summary	Optimisation of free-form building design is more challenging in terms of building information modelling and performance evaluation compared to conventional buildings. The paper provides a Photogrammetry-based BIM Modelling - Machine Learning Modelling - Multi-objective Optimisation framework to improve the solar energy utilisation potential of free-form buildings. Low altitude photogrammetry is used to collect the building and site environmental information. An ANN prediction model is developed using the control point coordinates and simulation data. Through parametric programming, the multi-objective algorithm is coupled with the ANN model to obtain the trade-off optimal building form. The results show that the maximum solar radiation value in winter can increase by 30.60% and the minimum solar radiation in summer can decrease by 13.99%. It is also shown that the integration of ANN modelling and photogrammetry-based BIM modelling into the multi-objective optimisation method can accelerate the optimisation process.
keywords	Multi-objective optimisation; Artificial neural network; Free-form shape building ; Solar energy utilisation
series	CAADRIA
email
last changed	2022/06/07 07:57

_id	sigradi2021_276
id	sigradi2021_276
authors	Álvarez, Marcelo and Bernal, Marcelo
year	2021
title	Design Spaces of Structurally Pre-evaluated Funicular Shapes
source	Gomez, P and Braida, F (eds.), Designing Possibilities - Proceedings of the XXV International Conference of the Ibero-American Society of Digital Graphics (SIGraDi 2021), Online, 8 - 12 November 2021, pp. 807–818
summary	In this paper, we develop a structural pre-evaluation and optimization technique for vault-like shapes. This implementation focuses on exploring design spaces in early design stages. The proposed technique approaches the problem of reducing the flexibility of the design space while advancing to later design stages for vault-like shapes. We start with a custom design space based on design intent. Then, we define a sampling criteria to study a reduced number of candidates. Later, the optimization process focuses on minimizing structural deformation values through shape manipulation. Results show a notorious enhancement for maximum deflection and displacement of the structure. Generally speaking, the shape optimization pattern is consistent with how vault-like shape works. All solutions reduce their span and boundary area while increasing the maximum height. Also, reaching maximum deformation values that are ten times better than the admissible final values on average.
keywords	Funicular shape, Structural optimization, Design space, Early-design stage, Particle-spring system
series	SIGraDi
email
last changed	2022/05/23 12:11

_id	ecaade2021_230
id	ecaade2021_230
authors	De Luca, Francesco, Sepúlveda, Abel and Varjas, Toivo
year	2021
title	Static Shading Optimization for Glare Control and Daylight
source	Stojakovic, V and Tepavcevic, B (eds.), Towards a new, configurable architecture - Proceedings of the 39th eCAADe Conference - Volume 2, University of Novi Sad, Novi Sad, Serbia, 8-10 September 2021, pp. 419-428
doi	https://doi.org/10.52842/conf.ecaade.2021.2.419
summary	Daylight and solar access influence positively building occupants' wellbeing and students' learning performance. However, an excess of sunlight can harm the visual comfort of occupants through disturbing glare effects. This study investigated, through multi-objective optimization, the potential of static shading devices to reduce glare and to guarantee daylight provision in a university building. The results showed that the reduction of disturbing glare was up to more than twice the reduced daylight, which nevertheless, was provided in adequate levels. View out and energy performance were also analyzed. Detailed results of optimal shading types and classrooms layout indications are presented.
keywords	Daylight; Visual comfort; Shading; Multi-objective optimization
series	eCAADe
email
last changed	2022/06/07 07:55

_id	ascaad2021_114
id	ascaad2021_114
authors	Houda, Maryam
year	2021
title	Materiality: Linking a Digital Material Framework with the Anthropological Hand
source	Abdelmohsen, S, El-Khouly, T, Mallasi, Z and Bennadji, A (eds.), Architecture in the Age of Disruptive Technologies: Transformations and Challenges [9th ASCAAD Conference Proceedings ISBN 978-1-907349-20-1] Cairo (Egypt) [Virtual Conference] 2-4 March 2021, pp. 568-580
summary	While computers and digital technology have evolved over the years and are changing the way we design and construct, some have criticized the way in which human tactility and intuition with material has diminished at the cost of increasing productivity and efficiency. Although the digital culture that architecture is engaged with today has brought about complex forms that could not have been possible by hand, there is a rising question of the place of craft and a hand-brain coordination in design, and the notion of learning through making. This paper explores the benefits and limitations of digital design tools in light of physically exploring building materials and gaining tactile intuition. While digital tools investigate structural optimisation methods using a parametric design workflow, physical experiments deal with understanding the transitional state of mud and its dynamic properties. This research is interested in how information is learnt from materiality during the physical act of making and what tactile experimentation can offer that the digital space cannot. Three key areas are explored: geometry and parametric variation, material properties and morphogenic behavior, as well as structural optimization methods using density grids. Force-matter relations are investigated through exploring material parameters through digital and physical form-finding processes as a way of exploring the notion of re-introducing the hand and craft in the design process which may bring about novel ways of thinking and doing.
series	ASCAAD
email
last changed	2021/08/09 13:13

_id	ijac202119313
id	ijac202119313
authors	Saldana Ochoa, Karla; Ohlbrock, Patrick Ole; D’Acunto, Pierluigi; Moosavi, Vahid
year	2021
title	Beyond typologies, beyond optimization: Exploring novel structural forms at the interface of human and machine intelligence
source	International Journal of Architectural Computing 2021, Vol. 19 - no. 3, 466–490
summary	This article presents a computer-aided design framework for the generation of non-standard structural forms in static equilibrium that takes advantage of the interaction between human and machine intelligence. The design framework relies on the implementation of a series of operations (generation, clustering, evaluation, selection, and regeneration) that allow to create multiple design options and to navigate in the design space according to objective and subjective criteria defined by the human designer. Through the interaction between human and machine intelligence, the machine can learn the nonlinear correlation between the design inputs and the design outputs preferred by the human designer and generate new options by itself. In addition, the machine can provide insights into the structural performance of the generated structural forms. Within the proposed framework, three main algorithms are used: Combinatorial Equilibrium Modeling for generating of structural forms in static equilibrium as design options, Self-Organizing Map for clustering the generated design options, and Gradient-Boosted Trees for classifying the design options. These algorithms are combined with the ability of human designers to evaluate non-quantifiable aspects of the design. To test the proposed framework in a real-world design scenario, the design of a stadium roof is presented as a case study.
keywords	Structural design, machine learning, topology, graphic statics, form-finding, Combinatorial Equilibrium Modeling, Self-Organizing Map, Gradient-Boosted Trees
series	journal
email
last changed	2024/04/17 14:29

_id	sigradi2021_140
id	sigradi2021_140
authors	Vivanco, Tomas and Yuan, Philip
year	2021
title	Robotic Weaving Manufacturing of Optimized Glass Fibres Panels
source	Gomez, P and Braida, F (eds.), Designing Possibilities - Proceedings of the XXV International Conference of the Ibero-American Society of Digital Graphics (SIGraDi 2021), Online, 8 - 12 November 2021, pp. 1235–1244
summary	This article presents the development of robotically fabricated and topological optimized fibreglass weaved panels. The process was led under the material intelligence workflow composed of the digital simulation of mechanical behaviour of the component, programming and optimization of the toolpath, and digital manufacturing with a common CNC machine. Through this process, the panels are optimized to minimize the use of material, decreasing the production time, to achieve its maximum mechanical and functional performance within its own design space.
keywords	machine learning, material computation, robotic fabrication
series	SIGraDi
email
last changed	2022/05/23 12:11

_id	ijac202119106
id	ijac202119106
authors	Del Campo, Matias; Alexandra Carlson, and Sandra Manninger
year	2021
title	Towards Hallucinating Machines - Designing with Computational Vision
source	International Journal of Architectural Computing 2021, Vol. 19 - no. 1, 88–103
summary	There are particular similarities in how machines learn about the nature of their environment, and how humans learn to process visual stimuli. Machine Learning (ML), more specifically Deep Neural network algorithms rely on expansive image databases and various training methods (supervised, unsupervised) to “make sense” out of the content of an image. Take for example how students of architecture learn to differentiate various architectural styles. Whether this be to differentiate between Gothic, Baroque or Modern Architecture, students are exposed to hundreds, or even thousands of images of the respective styles, while being trained by faculty to be able to differentiate between those styles. A reversal of the process, striving to produce imagery, instead of reading it and understanding its content, allows machine vision techniques to be utilized as a design methodology that profoundly interrogates aspects of agency and authorship in the presence of Artificial Intelligence in architecture design. This notion forms part of a larger conversation on the nature of human ingenuity operating within a posthuman design ecology. The inherent ability of Neural Networks to process large databases opens up the opportunity to sift through the enormous repositories of imagery generated by the architecture discipline through the ages in order to find novel and bespoke solutions to architectural problems. This article strives to demystify the romantic idea of individual artistic design choices in architecture by providing a glimpse under the hood of the inner workings of Neural Network processes, and thus the extent of their ability to inform architectural design.The approach takes cues from the language and methods employed by experts in Deep Learning such as Hallucinations, Dreaming, Style Transfer and Vision. The presented approach is the base for an in-depth exploration of its meaning as a cultural technique within the discipline. Culture in the extent of this article pertains to ideas such as the differentiation between symbolic and material cultures, in which symbols are defined as the common denominator of a specific group of people.1 The understanding and exchange of symbolic values is inherently connected to language and code, which ultimately form the ingrained texture of any form of coded environment, including the coded structure of Neural Networks.A first proof of concept project was devised by the authors in the form of the Robot Garden. What makes the Robot Garden a distinctively novel project is the motion from a purely two dimensional approach to designing with the aid of Neural Networks, to the exploration of 2D to 3D Neural Style Transfer methods in the design process.
keywords	Artificial intelligence, design agency, neural networks, machine learning, machine vision
series	journal
email
last changed	2021/06/03 23:29

_id	caadria2021_305
id	caadria2021_305
authors	Keshavarzi, Mohammad, Afolabi, Oladapo, Caldas, Luisa, Yang, Allen Y. and Zakhor, Avideh
year	2021
title	GenScan: A Generative Method for Populating Parametric 3D Scan Datasets
source	A. Globa, J. van Ameijde, A. Fingrut, N. Kim, T.T.S. Lo (eds.), PROJECTIONS - Proceedings of the 26th CAADRIA Conference - Volume 1, The Chinese University of Hong Kong and Online, Hong Kong, 29 March - 1 April 2021, pp. 91-100
doi	https://doi.org/10.52842/conf.caadria.2021.1.091
summary	The availability of rich 3D datasets corresponding to the geometrical complexity of the built environments is considered an ongoing challenge for 3D deep learning methodologies. To address this challenge, we introduce GenScan, a generative system that populates synthetic 3D scan datasets in a parametric fashion. The system takes an existing captured 3D scan as an input and outputs alternative variations of the building layout including walls, doors, and furniture with corresponding textures. GenScan is fully automated system that can also be manually controlled by a user through an assigned user interface. Our proposed system utilizes a combination of a hybrid deep neural network and a parametrizer module to extract and transform elements of a given 3D scan. GenScan takes advantage of style transfer techniques to generate new textures for the generated scenes. We believe our system would facilitate data augmentation to expand the currently limited 3D geometry datasets commonly used in 3D computer vision, generative design and general 3D deep learning tasks.
keywords	Computational Geometry; Generative Modeling; 3D Manipulation; Texture Synthesis
series	CAADRIA
email
last changed	2022/06/07 07:52

_id	caadria2021_308
id	caadria2021_308
authors	Wang, Dasong and Snooks, Roland
year	2021
title	Intuitive Behavior - The Operation of Reinforcement Learning in Generative Design Processes
source	A. Globa, J. van Ameijde, A. Fingrut, N. Kim, T.T.S. Lo (eds.), PROJECTIONS - Proceedings of the 26th CAADRIA Conference - Volume 1, The Chinese University of Hong Kong and Online, Hong Kong, 29 March - 1 April 2021, pp. 101-110
doi	https://doi.org/10.52842/conf.caadria.2021.1.101
summary	The paper posits a novel approach for augmenting existing generative design processes to embed a greater level of design intention and create more sophisticated generative methodologies. The research presented in the paper is part of a speculative research project, Artificial Agency, that explores the operation of Machine Learning (ML) in generative design and robotic fabrication processes. By framing the inherent limitation of contemporary generative design approaches, the paper speculates on a heuristic approach that hybridizes a Reinforcement Learning based top-down evolutionary approach with bottom-up emergent generative processes. This approach is developed through a design experiment that establishes a topological field with intuitive global awareness of pavilion-scale design criteria. Theoretical strategies and technical details are demonstrated in the design experiment in regard to the translation of ML definitions within a generative design context as well as the encoding of design intentions. Critical reflections are offered in regard to the impacts, characteristics, and challenges towards the further development of the approach. The paper attempts to broaden the range and impact of Artificial Intelligence applications in the architectural discipline.
keywords	Machine Learning; Generative Design Process; Multi-Agent Systems; Reinforcement Learning
series	CAADRIA
email
last changed	2022/06/07 07:58

_id	sigradi2021_146
id	sigradi2021_146
authors	Yönder, Veli Mustafa, Dogan, Fehmi and Çavka, Hasan Burak
year	2021
title	Deciphering and Forecasting Characteristics of Bodrum Houses Using Artificial Intelligence (AI) Approaches
source	Gomez, P and Braida, F (eds.), Designing Possibilities - Proceedings of the XXV International Conference of the Ibero-American Society of Digital Graphics (SIGraDi 2021), Online, 8 - 12 November 2021, pp. 241–252
summary	Computer vision (CV), artificial intelligence (AI), machine learning (ML), and deep learning (DL) applications, which are among the rapidly emerging and growing technologies, have the potential to be effectively used in the fields of architecture and construction. These applications are used not only in the field of architectural design development and construction site tracking but also to analyze and predict the architectural properties of existing buildings and heritage classification. This paper aims to classify and analyze the façades of Bodrum houses by using deep learning models, comprehensive relational database (RDB), and artificial neural network based clustering methods. Through the use of the above-mentioned methods, we managed to cluster Bodrum houses' façade attributes in five groups and testing image classification models in three different classifiers.
keywords	Image processing, Deep learning (DL), Classification, Hierarchical cluster analysis, Artificial neural networks (ANNs)
series	SIGraDi
email
last changed	2022/05/23 12:10

For more results click below: