• PDF / 2,090,255 Bytes
• 39 Pages / 439.37 x 666.142 pts Page_size
• 16 Downloads / 209 Views

DOWNLOAD

REPORT

---

CCDSF: A Computational Creative Design Systems Framework Jesús Pérez-Romero1

· Wendy Aguilar2

Received: 21 March 2020 / Accepted: 16 September 2020 / Published online: 23 November 2020 © Ohmsha, Ltd. and Springer Japan KK, part of Springer Nature 2020

Abstract From a mixed perspective of the fields of computational creativity and design research, we present a formal framework called Computational Creative Design Systems Framework (CCDSF) for describing, analyzing, and modeling Computational Creative Design Systems (CCDSs), either as autonomous agents or as collaborative systems integrating different agents, including both pure computational and human–computer scenarios. The CCDSF extends and unifies three existing frameworks: (i) the Function– Behavior–Structure framework for designing, to integrate design domain-specific knowledge; (ii) the Creative Systems Framework, to model design as a creative process; and (iii) the Conceptual Spaces Theory framework, to define the CCDSF at a more neutral level to be compatible with both the conceptual-space nature of the creative process and the implementation-level representation systems. The CCDSF core is a layered schema that structures the activity of designing through design concepts produced by design processes instrumented by agent-specific perception and action strategies, within a hierarchy of what we call design levels. Each layer of the schema constitutes a creative system that can implement exploratory, combinational, and transformational creativity mechanisms, including the possibility of defining the system at a meta-creative level. The hierarchical arrangement of creative systems defines the architecture of a CCDS. This architecture enables modeling CCDSs considering both the perspective of the generation and evaluation of artifacts that could be deemed creative and the precise modeling of the cognitive process of design, emphasizing its situated nature. To illustrate its usefulness, in the last part of this paper, we use the proposed framework to describe a real system in the videogame design domain.

B

Jesús Pérez-Romero [email protected] Wendy Aguilar [email protected]

1

Posgrado en Ciencia e Ingeniería de la Computación, Universidad Nacional Autónoma de México, Mexico City, Mexico

2

Department of Computer Science, Instituto de Investigaciones en Matemáticas Aplicadas y en Sistemas, Universidad Nacional Autónoma de México, 04510 Mexico City, Mexico

123

674

New Generation Computing (2020) 38:673–711

Keywords Computational creative design system · Designing schema · Artificial designer · Design computing

Introduction Since the emergence of the first computer aided design (CAD) tools in the mid1960s, the application of computing and artificial intelligence (mainly evolutionary computing techniques) to the design domain has been focused mostly on the modeling of design objects and their optimization on specific performance criteria, supported by analytical methods, computational simulations, and human evaluation. However, these processes correspon