Abstract

The following paper investigates the idea of reducing the human digital intervention to a minimum during the advanced design process. Augmenting the outcome attributes beyond the designer's capabilities by computational design methods, data collection, data computing and digital fabrication, altogether imitating the human design process. The primary technical goal of the research was verification of restrictions and abilities used in examined hardware and software tools: data collection tools, Machine Learning algorithms and computer-aided manufacturing. The generated effect became the self-interpretation of intentions signalised by the human-designer. The interpretation was possible with Artificial Neural Networks training based on learning sound samples and tracked designer movement in the act of drawing. Experiments led to design methods, in which primary focus is set on machine interpreting intentions of the user/designer than executing just precise instructions. That approach gives more space for possible solutions, and at the same time leaves the room for whole kinds of form optimisation. Moreover, the process of interpretation may be self-developing in time, as Artificial Neural Network may be trained more for each of the work she will execute. The following paper presents the workflow for both visual and sound data capturing along with the data processing, creating the methods of the human design process emulation. Additionally, the research presents the possible implementation of machine learning in the design process. Presented approach reinterprets theoretical roles in the human-machine dialogue during the creative process in the general field of computational design. The research aimed to preserve designers individual preferences, knowledge and skills, to be processed by an advanced, digital entity, resulting in machine-made forms.

Citations

Authors (2)