Open AccessApplication of generative design methods using multimodal data in the field of architecture and urban planning
Author(s) -
Anastasia Laushkina,
Oleg Basov
Publication year - 2021
Publication title -
naučnyj rezulʹtat. informacionnye tehnologii
Language(s) - English
Resource type - Journals
ISSN - 2518-1092
DOI - 10.18413/2518-1092-2021-6-3-0-1
Subject(s) - generative design , generative grammar , computer science , field (mathematics) , process (computing) , set (abstract data type) , architecture , engineering design process , generative model , mode (computer interface) , conceptual design , iterative and incremental development , systems engineering , artificial intelligence , industrial engineering , software engineering , human–computer interaction , engineering , mathematics , mechanical engineering , art , metric (unit) , operations management , pure mathematics , visual arts , programming language , operating system
This paper is devoted to the study of generative modeling in design used: this is an iterative process that generates several output data with consideration to certain restrictions of the input parameters. Generative design can offer advantages to traditional design as specifically defined criteria are taken into account early as the conceptual design stage. The purpose of the work was to study the existing methods and algorithms of generative design in urban planning, taking into account changes in environmental characteristics. The first step was to determine the input data for the design of the urban structure. The second stage was the analysis of existing solutions. In conclusion, it was The conclusion was made about the various constraints faced by developers, namely: the requirement to make changes in the user mode imposes restrictions on the use of such systems. It is also established that at the moment there is no clear logic by which generative systems can identify contradictions in a set of multimodal data and make a decision on the correspondence of the obtained result to dynamically changing parameters.