Browsing by Author "Piskorec, Luka"

Now showing 1 - 6 of 6
  • No Thumbnail Available
    Item
    CURVED.IT: A design tool to integrate making with curved folding into digital design process
    (Orhan Hacihasanoglu ITU Faculty of Architecture, 2019-05) Bacinoglu, Saadet; Piskorec, Luka; Kotnik, Toni; Department of Architecture
    The act of changing the direction of a sheet surface along a non-straight curve is a specific case of curved folding. From an architectural point of view, curved folding is an exciting operation. One or a couple of operation can generate highly complex shell-like spatial enclosure. From a digital design perspective, the implementation of curved folding with the built-in toolsets of available computer-aided design softwares is a challenging problem. The equilibrium state of curved folded geometry is needed to be found with a computational form-finding strategy. To use curved folding as a digital design operation, we introduce a new tool through developing a digital procedure for form-finding. The tool we develop can enable the experimentation with curved folding in the early stage of design process and facilitate the subsequent design development. In this article, we briefly present the literature focusing on curved folding in computational geometry, as well as the scope and description of a subclass of curved folding operation. Then, we introduce a digital tool, CURVED.IT through a design manual for its implementation and an algorithmic framework for its extension. Lastly, we discuss the design examples generated by CURVED.IT, and the potentials of the tool.
  • Thumbnail Image
    Item
    Deep Form Finding Using Variational Autoencoders for deep form finding of structural typologies
    (2019) de Miguel, Jaime; Eugenia Villafane, Maria; Piskorec, Luka; Sancho-Caparrini, Fernando; Department of Architecture; Sousa, JP; Henriques, GC; Xavier, JP; University of Seville; University College London
    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.
  • Thumbnail Image
    Item
    An insight to computation: A study of the potentials of computation as well as digital fabrication in architectural design
    (2021) Hautala, Aino; Piskorec, Luka; ark; Taiteiden ja suunnittelun korkeakoulu; School of Arts, Design and Architecture; Kotnik, Toni
    Computational design is often misunderstood in the architectural profession. It is seen as an enabler of monumental design, detached from the realities most architects have to adhere to and considered too complicated to be of actual use for everyday design. This, however, is a misrepresentation. This thesis aims to showcase the potentials of computation in creating not only complex design but better design. The focus of the thesis is on understanding what computation is, how it can aid design and how it will change the way we practice architecture. This is achieved by referencing multiple renowned experts and by introducing many inspiring projects that showcase computational potential in a wide variety of ways. The thesis also examines the relationship between computational design, digital data and digital fabrication. The thesis starts with an explanation of what computation is and how it works. Part 1: Context for computation also examines why the development of computation was so significant for architecture and how it has evolved throughout the years. In Part 2: Realising computation, the discussion becomes more practical. This chapter introduces various approaches for the utilisation of computation and methods of digital fabrication. Part 3: Design of the future focuses on the benefits of computation, especially from the environmental point of view. Here are also expressed thoughts on the issues computation needs to face before breaking through to the mainstream. Part 4 contains selected case studies that are especially interesting from a computational point of view. The hope is that this thesis will make readers more aware of computation and perhaps inspire some to experiment with it independently. It is not as complicated as it might initially appear.
  • Thumbnail Image
    Item
    Integrating computational planting design to the landscape design process
    (2021) Luminiitty, Rosaliina; Piskorec, Luka; ark; Taiteiden ja suunnittelun korkeakoulu; School of Arts, Design and Architecture; Fricker, Pia
    As urbanization continues, there is an ever-growing need to support biodiversity in the urban areas with solutions that provide high functionality with low maintenance requirements. Naturalistic planting design principles provide the means to design ecologically informed planting areas with high aesthetical values. The complexity of the naturalistic plantings, with multi-layered vegetation and diverse planting principles, is the key to enhancing biodiversity in the urban environment. However, from the planting design's workflow perspective, the complexity has brought up the request for more informed sketching and decision-making tools. This thesis explores how the integration of computational planting design can promote the biodiversity of the herbaceous planting design. Through Grasshopper, a visual algorithm editor software plugin for the Rhinoceros, the aim is to explore how computational planting de-sign could be implemented as part of the design workflow's early stages. This thesis aims to identify how computational design, ecology, and naturalistic planting design discuss complexity and patterns and how these are all related to biodiversity. The thesis consists of four parts. The theoretical part discusses relevant literature and case studies to comprehend how computational design, data, and computational planting design are connected to biodiversity. In addition, biodiversity, ecology, naturalistic planting, and their patterns were discussed. The aim is to explain why computational design is relevant when biodiversity loss is considered. The second part examines the current planting design work-flow and identifies the factors related to plant placement and its principles from an ecological perspective. The third practical part presents the author's process to implement the computational design into sketching workflow and aims to identify design strategies for plant placement and area divisions. In conclusion, it can be conducted that computational planting design will improve biodiversity in the early phases of the planting design workflow. The computational planting design will provide explorative and iterative planting design possibilities, especially in the sketching phase. The computational design will ease the design process and increase the manageability of diverse, species-rich, complex plantings. Moreover, as analysis and information will be included as part of the design process, this will enable up-to-date analysis and informed decision-making towards a more resilient future.
  • No Thumbnail Available
    Item
    The scaleless and the bodily: Formal explorations of abstract spatial relationships
    (2021) Kleinbergs, Kristaps; Piskorec, Luka; ark; Taiteiden ja suunnittelun korkeakoulu; School of Arts, Design and Architecture; Kotnik, Toni
  • No Thumbnail Available
    Item
    Structuralism: Patterns of Interaction Computational design thinking across scales
    (Wichmann Verlag im VDE Verlag GmbH, 2019-05-21) Fricker, Pia; Kotnik, Toni; Piskorec, Luka; Department of Architecture
    Digital design is driven by a thinking in structures that emphasizes patterns of order and their interaction. The paper presents an introduction into such design thinking and emphasizes the exploration of organizational pattern in various level of abstraction and scale through the integration of computational methods and workflows. Using a speculative design studio, held at Aalto University as a case study, the paper introduces novel ways of integrating patterns as relational frameworks, in order to formulate future-oriented answer to complex design challenges focused on landscape and urban questions. Special focus is set on the discussion of the integration of theory based computational methods within a teaching environment.