Towards machine learning for architectural fabrication in the age of industry 4.0

Abstract

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.

[Retracted] Dynamic Simulation Modeling of Industrial Robot Kinematics in Industry 4.0

“…Change is a constant truth, and Industry 4.0 also poses new challenges to industrial robots in the Industry 4.0 environment [9]. Ramsgaard omsen et al first added a parameter β i about y-axis rotation to the D-H model to avoid the singularity problem of parallel joints of industrial robots and established a new modified MD-H motion model based on the D-H model [10]. Based on the D-H model, Chauhan and Khare added two parameters to construct a six-parameter S model to better express the motion relationship of each linkage of the industrial robot [11].…”

Leading-Edge Technologies for Architectural Design: A Comprehensive Review

“…With a parametric model in place, architects can effortlessly manipulate these variables, facilitating real-time adjustments and immediate feedback on the design's performance. The integration of machine learning enriches generative design by enhancing the system's adaptive capabilities (Wibranek and Tessmann, 2021;Ramsgaard et al, 2020;Yazici, 2020;Ampanavos et al, 2021;Warnett and Zdun, 2022;Meekings and Schnabel, 2017). Machine learning algorithms analyze extensive datasets encompassing architectural precedents, construction methods, and environmental conditions.…”

Robotic plaster spraying – Crafting surfaces with adaptive thin-layer printing

“…However, in recent years, research in digital fabrication has aimed to tackle the issue of the prohibitive complexity of simulation by developing data-driven design tools to explore the design potentials of specific processes with a relative precision in the prediction, driven by sensor systems that are becoming more precise and easily accessible. [15][16][17][18] One current example in that regard is the research project Spatial Wire Cutting, 19 where the complexity for the prediction of the fabrication process comes from the respective interaction of a loose and form-adaptive hot-wire adapting itself against the resistance of the processed material. In this research project, data from the fabrication parameters such as heat input, cutting speed, and forces are aligned with the resulting geometries.…”