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 624

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

_id	ecaadesigradi2019_498
id	ecaadesigradi2019_498
authors	Bermek, Mehmet Sinan, Shelden, Dennis and Gentry, T. Russel
year	2019
title	A Holistic Approach to Feature-based Structural Mapping in Cross Laminated Timber Buildings
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 2, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 789-796
doi	https://doi.org/10.52842/conf.ecaade.2019.2.789
summary	Mass Engineered Timber products provide a unique opportunity in configuring panelized building systems that are suitable for both prefabrication and onsite customization. The structural nature of these large section elements also brings about the need for a coordinated design-fabrication-assembly workflow. These products can assume different geometric configurations and their behaviour can be approximated globally by simplifying framing schemas. Current BIM Interoperability standards such as STEP or IFC already acknowledge and support the interconnected nature of component properties, yet these Data Models are component focused. Expanding on the relationships between components and using sets to define part to whole, or exteriority relationships could yield a more flexible and agile querying of building information.This would be a framework fit for automated feature derivation and rule based design applications. To this end Graph structures and Graph Databases, alongside existing ontology authoring tools are studied to probe new cognitive possibilities in collaborative AEC workflows
keywords	Graph theory; BIM; CLT; IFC
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:52

_id	ecaadesigradi2019_453
id	ecaadesigradi2019_453
authors	Nisztuk, Maciej and Myszkowski, Pawe³
year	2019
title	Tool for evolutionary aided architectural design. Hybrid Evolutionary Algorithm applied to Multi-Objective Automated Floor Plan Generation
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 1, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 61-70
doi	https://doi.org/10.52842/conf.ecaade.2019.1.061
summary	The paper presents the ELISi multi-criteria optimisation application for AFPG based on Hybrid Evolutionary Algorithm (HEA). The research aims to create functional computational design tool for architects, mimicking the workflow of architectural design process. The article includes explanation of the proposed approach: problem representation, genetic algorithm operators, fitness functions definitions, post processing operations, software functionalities and workflow as well as achieved architectural results and outline of future research.
keywords	MOO in CAAD; automated floor plan generation; HEA
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:58

_id	ecaadesigradi2019_602
id	ecaadesigradi2019_602
authors	Toulkeridou, Varvara
year	2019
title	Steps towards AI augmented parametric modeling systems for supporting design exploration
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 1, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 81-90
doi	https://doi.org/10.52842/conf.ecaade.2019.1.081
summary	Dataflow parametric modeling environments have become popular as exploratory tools due to them allowing the variational exploration of a design by controlling the parameters of its parametric model schema. However, the nature of these systems requires designers to prematurely commit to a structure and hierarchy of geometric relationships, which makes them inflexible when it comes to design exploration that requires topological changes to the parametric modeling graph. This paper is a first step towards augmenting parametric modeling systems via the use of machine learning for assisting the user towards topological exploration. In particular, this paper describes an approach where Long Short-Term Memory recurrent neural networks, trained on a data set of parametric modeling graphs, are used as generative systems for suggesting alternative dataflow graph paths to the parametric model under development.
keywords	design exploration; visual programming; machine learning
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:58

_id	acadia19_168
id	acadia19_168
authors	Adilenidou, Yota; Ahmed, Zeeshan Yunus; Freek, Bos; Colletti, Marjan
year	2019
title	Unprintable Forms
source	ACADIA 19:UBIQUITY AND AUTONOMY [Proceedings of the 39th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-59179-7] (The University of Texas at Austin School of Architecture, Austin, Texas 21-26 October, 2019) pp.168-177
doi	https://doi.org/10.52842/conf.acadia.2019.168
summary	This paper presents a 3D Concrete Printing (3DCP) experiment at the full scale of virtualarchitectural bodies developed through a computational technique based on the use of Cellular Automata (CA). The theoretical concept behind this technique is the decoding of errors in form generation and the invention of a process that would recreate the errors as a response to optimization (Adilenidou 2015). The generative design process established a family of structural and formal elements whose proliferation is guided through sets of differential grids (multi-grids) leading to the build-up of large span structures and edifices, for example, a cathedral. This tooling system is capable of producing, with specific inputs, a large number of outcomes in different scales. However, the resulting virtual surfaces could be considered as "unprintable" either due to their need of extra support or due to the presence of many cavities in the surface topology. The above characteristics could be categorized as errors, malfunctions, or undesired details in the geometry of a form that would need to be eliminated to prepare it for printing. This research project attempts to transform these "fabrication imprecisions" through new 3DCP techniques into factors of robustness of the resulting structure. The process includes the elimination of the detail / "errors" of the surface and their later reinsertion as structural folds that would strengthen the assembly. Through this process, the tangible outputs achieved fulfill design and functional requirements without compromising their structural integrity due to the manufacturing constraints.
series	ACADIA
type	normal paper
email
last changed	2022/06/07 07:54

_id	caadria2019_660
id	caadria2019_660
authors	Aghaei Meibodi, Mania, Giesecke, Rena and Dillenburger, Benjamin
year	2019
title	3D Printing Sand Molds for Casting Bespoke Metal Connections - Digital Metal: Additive Manufacturing for Cast Metal Joints in Architecture
source	M. Haeusler, M. A. Schnabel, T. Fukuda (eds.), Intelligent & Informed - Proceedings of the 24th CAADRIA Conference - Volume 1, Victoria University of Wellington, Wellington, New Zealand, 15-18 April 2019, pp. 133-142
doi	https://doi.org/10.52842/conf.caadria.2019.1.133
summary	Metal joints play a relevant role in space frame constructions, being responsible for large amount of the overall material and fabrication cost. Space frames which are constructed with standardized metal joints are constrained to repetitive structures and topologies. For customized space frames, the fabrication of individual metal joints still remains a challenge. Traditional fabrication methods such as sand casting are labour intensive, while direct 3D metal printing is too expensive and slow for the large volumes needed in architecture.This research investigates the use of Binder Jetting technology to 3D print sand molds for casting bespoke metal joints in architecture. Using this approach, a large number of custom metal joints can be fabricated economically in short time. By automating the generation of the joint geometry and the corresponding mold system, an efficient digital process chain from design to fabrication is established. Several design studies for cast metal joints are presented. The approach is successfully tested on the example of a full scale space frame structure incorporating almost two hundred custom aluminum joints.
keywords	3D printing; binder jetting; sand casting; metal joints; metal casting; space frame; digital fabrication; computational design; lightweight; customization
series	CAADRIA
email
last changed	2022/06/07 07:54

_id	ecaadesigradi2019_605
id	ecaadesigradi2019_605
authors	Andrade Zandavali, Bárbara and Jiménez García, Manuel
year	2019
title	Automated Brick Pattern Generator for Robotic Assembly using Machine Learning and Images
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 3, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 217-226
doi	https://doi.org/10.52842/conf.ecaade.2019.3.217
summary	Brickwork is the oldest construction method still in use. Digital technologies, in turn, enabled new methods of representation and automation for bricklaying. While automation explored different approaches, representation was limited to declarative methods, as parametric filling algorithms. Alternatively, this work proposes a framework for automated brickwork using a machine learning model based on image-to-image translation (Conditional Generative Adversarial Networks). The framework consists of creating a dataset, training a model for each bond, and converting the output images into vectorial data for robotic assembly. Criteria such as: reaching wall boundary accuracy, avoidance of unsupported bricks, and brick's position accuracy were individually evaluated for each bond. The results demonstrate that the proposed framework fulfils boundary filling and respects overall bonding structural rules. Size accuracy demonstrated inferior performance for the scale tested. The association of this method with 'self-calibrating' robots could overcome this problem and be easily implemented for on-site.
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:54

_id	ecaadesigradi2019_290
id	ecaadesigradi2019_290
authors	Assem, Ayman, Abdelmohsen, Sherif and Ezzeldin, Mohamed
year	2019
title	A Fuzzy-Based Approach for Evaluating Existing Spatial Layout Configurations
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 2, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 35-44
doi	https://doi.org/10.52842/conf.ecaade.2019.2.035
summary	This paper proposes a fuzzy-based approach for the automated evaluation of spatial layout configurations. Our objective is to evaluate soft and interdependent design qualities (such as connectedness, enclosure, spaciousness, continuity, adjacency, etc.), to satisfy multiple and mutually inclusive criteria, and to account for all potential and logical solutions without discarding preferable, likely or even less likely possible solutions. Using fuzzyTECH, a fuzzy logic software development tool, we devise all possible spatial relation inputs affecting physical and non-physical outputs for a given space using descriptive rule blocks. We implement this fuzzy logic system on an existing residential space to evaluate different layout alternatives. We define all linguistic input variables, output variables, and fuzzy sets, and present space-space relations using membership functions. We use the resulting database of fuzzy agents to evaluate the design of the existing residential spaces.
keywords	Fuzzy logic; Space layout planning; Heuristic methods
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:54

_id	acadia19_258
id	acadia19_258
authors	Bar-Sinai, Karen Lee; Shaked, Tom; Sprecher, Aaron
year	2019
title	Informing Grounds
source	ACADIA 19:UBIQUITY AND AUTONOMY [Proceedings of the 39th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-59179-7] (The University of Texas at Austin School of Architecture, Austin, Texas 21-26 October, 2019) pp. 258-265
doi	https://doi.org/10.52842/conf.acadia.2019.258
summary	Advancements in robotic fabrication are enabling on-site construction in increasingly larger scales. In this paper, we argue that as autonomous tools encounter the territorial scale, they open new ways to embed information into it. To define the new practice, this paper introduces a protocol combining a theoretical framework and an iterative process titled Informing Grounds. This protocol mediates and supports the exchange of knowledge between a digital and a physical environment and is applicable to a variety of materials with uncertain characteristics in a robotic manufacturing scenario. The process is applied on soil and demonstrated through a recent design-to-fabrication workshop that focused on simulating digital groundscaping of distant lunar grounds employing robotic sand-forming. The first stage is ‘sampling’—observing the physical domain both as an initial step as well as a step between the forming cycles to update the virtual model. The second stage is ‘streaming’—the generation of information derived from the digital model and its projection onto the physical realm. The third stage is ‘transforming’—the shaping of the sand medium through a physical gesture. The workshop outcomes serve as the basis for discussion regarding the challenges posed by applying autonomous robotic tools on materials with uncertain behavior at a large-scale.
series	ACADIA
type	normal paper
email
last changed	2022/06/07 07:54

_id	caadria2019_195
id	caadria2019_195
authors	Brandao, Filipe JS and Paio, Alexandra
year	2019
title	Context-Aware Mass Customization Construction System - Methods for user captured as-built plans
source	M. Haeusler, M. A. Schnabel, T. Fukuda (eds.), Intelligent & Informed - Proceedings of the 24th CAADRIA Conference - Volume 2, Victoria University of Wellington, Wellington, New Zealand, 15-18 April 2019, pp. 101-110
doi	https://doi.org/10.52842/conf.caadria.2019.2.101
summary	The problem of context, a fundamental aspect of dealing with built environments, has not been adequately addressed by mass customization systems so far, which has limited their scope of application. The aim of the present article is to evaluate the adequacy of existing methods of producing as-built plans of rooms by non-expert users for the automatic generation and production of partition walls for building renovation. This paper highlights criteria to develop appropriate methods of capturing context for mass customization construction systems.
keywords	Mass Customization; As-built Plans; Building Renovation; Computational design
series	CAADRIA
email
last changed	2022/06/07 07:54

_id	ecaadesigradi2019_473
id	ecaadesigradi2019_473
authors	Brandao, Filipe, Paio, Alexandra and Lopes, Adriano
year	2019
title	Interactive algorithm for generating accurate as-built plans by building owners
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 2, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 69-78
doi	https://doi.org/10.52842/conf.ecaade.2019.2.069
summary	Mass Customization systems in architecture have yet to adequately address the problem of capturing physical context, a fundamental aspect of dealing with building renovation, which has limited their scope of application. Previous research has demonstrated that existing methods of capturing as-built plans of rooms by non-expert users do not produce sufficiently accurate results for digital fabrication. The present paper reports on research into the development of an algorithm for semi-automated survey of convex or non-convex rooms by building owners. The improved workflow is tested by expert and non-expert users in a to-be renovated building and the results are compared with existing methods of survey.
keywords	Mass Customization; As-built Plans; Building renovation; Polygon partition;
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:54

_id	ijac201917106
id	ijac201917106
authors	Brown, Nathan C. and Caitlin T. Mueller
year	2019
title	Design variable analysis and generation for performance-based parametric modeling in architecture
source	International Journal of Architectural Computing vol. 17 - no. 1, 36-52
summary	Many architectural designers recognize the potential of parametric models as a worthwhile approach to performance- driven design. A variety of performance simulations are now possible within computational design environments, and the framework of design space exploration allows users to generate and navigate various possibilities while considering both qualitative and quantitative feedback. At the same time, it can be difficult to formulate a parametric design space in a way that leads to compelling solutions and does not limit flexibility. This article proposes and tests the extension of machine learning and data analysis techniques to early problem setup in order to interrogate, modify, relate, transform, and automatically generate design variables for architectural investigations. Through analysis of two case studies involving structure and daylight, this article demonstrates initial workflows for determining variable importance, finding overall control sliders that relate directly to performance and automatically generating meaningful variables for specific typologies.
keywords	Parametric design, design space formulation, data analysis, design variables, dimensionality reduction
series	journal
email
last changed	2019/08/07 14:04

_id	caadria2019_491
id	caadria2019_491
authors	Cai, Chenyi, Tang, Peng and Li, Biao
year	2019
title	Intelligent Generation of Architectural layout inheriting spatial features of Chinese Garden Based on Prototype and Multi-agent System - A Case Study on Lotus Teahouse in Yixing
source	M. Haeusler, M. A. Schnabel, T. Fukuda (eds.), Intelligent & Informed - Proceedings of the 24th CAADRIA Conference - Volume 1, Victoria University of Wellington, Wellington, New Zealand, 15-18 April 2019, pp. 291-300
doi	https://doi.org/10.52842/conf.caadria.2019.1.291
summary	This study presents an approach for the intelligent generation of architectural layout, in which partial space inherits Chinese garden spatial features. The approach combines spatial prototype analysis and evolutionary optimization process. On one hand, from the perspective of shape grammar, this paper both analyzes and abstracts the spatial prototype that describes the spatial characteristics of Chinese gardens, including the organization system of architecture and landscape, with the spatial sequences along the tourism orientation. On the other hand, taking the design task of Lotus teahouse as an example, a typical spatial prototype is selected to develop the generative intelligent experiment to achieve the architectural layout, in which the spatial prototype is inherited. Through rule-making and parameter adjustment, the spatial prototype will eventually be transformed into a computational model based on the multi-agent system. Hence, the experiment of intelligent generation of architectural layout is carried out under the influence of the function, form and environmental factors; and a three-dimensional conceptual model that inherits the Chinese garden spatial prototype is obtained ultimately.
keywords	Chinese garden; Architectural layout; Spatial prototype; Multi-agent system; Intelligent generation
series	CAADRIA
email
last changed	2022/06/07 07:54

_id	ecaadesigradi2019_210
id	ecaadesigradi2019_210
authors	Castriotto, Caio, Giantini, Guilherme and Celani, Gabriela
year	2019
title	Biomimetic Reciprocal Frames - A design investigation on bird's nests and spatial structures
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 1, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 613-620
doi	https://doi.org/10.52842/conf.ecaade.2019.1.613
summary	Reciprocal Frame (RF) is a constructive system typically applied with timber, since it is composed by discrete elements with short dimensions. It allows the construction of large spans and complex geometries. This kind of structure has been addressed by recent research projects that aim to produce it using computational tools and digital fabrication techniques. Moreover, the enhancement of these technologies enabled the integration of simulations of biological processes into the design process as a way to obtain better and optimal results, which is known as Biomimetics. This paper describes the development of a spatial structure that combines the principles of RF and the assembly process of natural agents, such as birds, in a digital environment. The tools used for the generation of the structure were Rhinoceros, Grasshopper and different add-ons, such as Culebra, Kangaroo, Pufferfish and Weaverbird.
keywords	Biomimetics; Reciprocal Frame; Nexorade; Computational Design; Agent-Based System
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:55

_id	acadia19_564
id	acadia19_564
authors	Chai, Hua; Marino, Dario; So, ChunPong; Yuan, Philip F.
year	2019
title	Design for Mass-Customization
source	ACADIA 19:UBIQUITY AND AUTONOMY [Proceedings of the 39th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-59179-7] (The University of Texas at Austin School of Architecture, Austin, Texas 21-26 October, 2019) pp. 564-572
doi	https://doi.org/10.52842/conf.acadia.2019.564
summary	Tradition wood tectonics, like interlocking joints, have regained focus against the background of digital design and fabrication technologies. While research on interlocking joints is quite focused on joint geometries, especially for timber plates, there has been less attention on the design and mass customization of interlocking joints for linear timber elements. In this context, this research addresses the challenges of mass customization of interlocking joints for linear elements through the design and realization of a 9-meterhigh timber structure with fully interlocking joints, without the use of any nails or glue. A customized code generation program was developed for the fabrication process, allowing the rapid programming and fabrication for all the 840 elements and 2592 notches. The project demonstrates how innovative structures are allowed through the synthesis of joint geometry, assembly process, and cutting-edge fabrication technology.
series	ACADIA
type	normal paper
email
last changed	2022/06/07 07:55

_id	cf2019_021
id	cf2019_021
authors	Cheng, Chi-Li and June-Hao Hou
year	2019
title	A Method of Mesh Simplification for Drone 3D Modeling with Architectural Feature Extraction
source	Ji-Hyun Lee (Eds.) "Hello, Culture!" [18th International Conference, CAAD Futures 2019, Proceedings / ISBN 978-89-89453-05-5] Daejeon, Korea, p. 169
summary	This paper proposes a method of mesh simplification for 3D terrain or city models generated photogrammetrically from drone captured images, enabled by the ability of extracting the architectural features. Compare to traditional geometric computational method, the proposed method recognizes and processes the features from the architectural perspectives. In addition, the workflow also allows exporting the simplified models and geometric features to open platforms, e.g. OpenStreetMap, for practical usages in site analysis, city generation, and contributing to the open data communities.
keywords	Mesh Reconstruction, photogrammetry, mesh simplification, procedural mode, machine learning
series	CAAD Futures
email
last changed	2019/07/29 14:08

_id	acadia19_642
id	acadia19_642
authors	Chua, Pamela Dychengbeng; Hui, Lee Fu
year	2019
title	Compliant Laminar Assemblies
source	ACADIA 19:UBIQUITY AND AUTONOMY [Proceedings of the 39th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-59179-7] (The University of Texas at Austin School of Architecture, Austin, Texas 21-26 October, 2019) pp. 642-653
doi	https://doi.org/10.52842/conf.acadia.2019.642
summary	This paper presents an innovative approach to the design and fabrication of three-dimensional objects from single-piece flat sheets, inspired by the origami technique of twist-closing. While in origami twist-closing is merely used to stabilize a cylindrical or spherical structure, ensuring it maintains its shape, this research investigates the potential of twist-closing as a multi-functional mechanism that also activates and controls the transformation of a planar surface into a predesigned three-dimensional form. This exploration is directed towards an intended application to stiff and brittle sheet materials that are difficult to shape through other processes. The methods we have developed draw mainly upon principles of lattice kirigami and laminar reciprocal structures. These are reflected in a workflow that integrates digital form-generation and fabrication-rationalization techniques to reference and apply these principles at every stage. Significant capabilities of the developed methodology include: (1) achievement of pseudo-double-curvature with brittle, stiff sheet materials; (2) stabilization in a 3D end-state as a frameless self-contained single-element laminar reciprocal structure—essentially a compliant mechanism; and (3) an ability to pre-encode 3D assembly constraints in a 2D cutout pattern, which guides a moldless fabrication process. The paper reviews the precedent geometric techniques and principles that comprise this method of 3D surface fabrication and describes a sample deployment of the method as applied to the design of laminar modules made of high-pressure laminate (HPL).
series	ACADIA
type	normal paper
email
last changed	2022/06/07 07:54

_id	ecaadesigradi2019_404
id	ecaadesigradi2019_404
authors	Collins, Jeffrey and Gentry, Russell
year	2019
title	Coordinating Atypical Architectural Precast Concrete Façades - Two categories
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 2, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 261-268
doi	https://doi.org/10.52842/conf.ecaade.2019.2.261
summary	This research focuses on issues of coordination between designers and fabricators during early design. The aim of this work is to improve representations, enable more informed conversations, and streamline exchanges of digital models. In order to show the potential of the work, research is focused on architectural precast concrete facades. Previous work established methods for linking "global" and "local" parametric models of architectural intent and corresponding components, describing processes of mapping from individual custom panels to diagrammatic façade surfaces and vice-versa. Such mapping may be considered "direct," wherein individual panel boundaries - defined by surface patterning - allow simple mapping of data from global to local or from local to global descriptions. However, there are some buildings with architectural precast concrete façades which do not permit direct relationships between global and local descriptions. These atypical facades require "indirect" maps containing additional layers of information in order to coordinate global and local descriptions. This paper describes two categories of these indirect scenarios: "panelization" and "patterns across panels."
keywords	BIM; Parametric modelling; Architectural precast concrete; Building facades
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:56

_id	ecaadesigradi2019_514
id	ecaadesigradi2019_514
authors	de Miguel, Jaime, Villafa?e, Maria Eugenia, Piškorec, Luka and Sancho-Caparrini, Fernando
year	2019
title	Deep Form Finding - Using Variational Autoencoders for deep form finding of structural typologies
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 1, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 71-80
doi	https://doi.org/10.52842/conf.ecaade.2019.1.071
summary	In this paper, we are aiming to present a methodology for generation, manipulation and form finding of structural typologies using variational autoencoders, a machine learning model based on neural networks. We are giving a detailed description of the neural network architecture used as well as the data representation based on the concept of a 3D-canvas with voxelized wireframes. In this 3D-canvas, the input geometry of the building typologies is represented through their connectivity map and subsequently augmented to increase the size of the training set. Our variational autoencoder model then learns a continuous latent distribution of the input data from which we can sample to generate new geometry instances, essentially hybrids of the initial input geometries. Finally, we present the results of these computational experiments and lay out the conclusions as well as outlook for future research in this field.
keywords	artificial intelligence; deep neural networks; variational autoencoders; generative design; form finding; structural design
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:55

_id	cf2019_051
id	cf2019_051
authors	Dickey , Rachel
year	2019
title	Soft Additive Fabrication Processes: Material Indeterminacy in 3D Printing
source	Ji-Hyun Lee (Eds.) "Hello, Culture!" [18th International Conference, CAAD Futures 2019, Proceedings / ISBN 978-89-89453-05-5] Daejeon, Korea, p. 434
summary	This description of Soft Additive Fabrication Processes, documents ways in which chance and randomness might be treated as values rather than problems. The production of a series of robotically controlled extruder experiments explore integrating material volition with the rigid order of machine control. Specifically this paper outlines the development of tooling procedures that harness emergent conditions in the automation of qualitative material effects. A key question for the research asks, how might architects imagine a design and construction scenario, which is no longer confined to prescriptive material dimensions, but is instead driven by digitally calibrated stochastic material processes? What opportunities might arise from developing an automated system, which does not rely on direct translation, but instead operates and predicts outcomes within a range of potential results?
keywords	Additive manufacturing, robotics, 3D printing, indeterminacy, material volition
series	CAAD Futures
email
last changed	2019/07/29 14:18

_id	ecaadesigradi2019_648
id	ecaadesigradi2019_648
authors	Eisenstadt, Viktor, Langenhan, Christoph and Althoff, Klaus-Dieter
year	2019
title	Generation of Floor Plan Variations with Convolutional Neural Networks and Case-based Reasoning - An approach for transformative adaptation of room configurations within a framework for support of early conceptual design phases
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 2, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 79-84
doi	https://doi.org/10.52842/conf.ecaade.2019.2.079
summary	We present an approach for computer-aided generation of different variations of floor plans during the early phases of conceptual design in architecture. The early design phases are mostly characterized by the processes of inspiration gaining and search for contextual help in order to improve the building design at hand. The generation method described in this work uses the novel as well as established artificial intelligence methods, namely, generative adversarial nets and case-based reasoning, for creation of possible evolutions of the current design based on the most similar previous designs. The main goal of this approach is to provide the designer with information on how the current floor plan can evolve over time in order to influence the direction of the design process. The work described in this paper is part of the methodology FLEA (Find, Learn, Explain, Adapt) whose task is to provide a holistic structure for support of the early conceptual phases in architecture. The approach is implemented as the adaptation component of the framework MetisCBR that is based on FLEA.
keywords	room configuration; adaptation; case-based reasoning; convolutional neural networks; conceptual design
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:55

For more results click below: