Shuichi Fukuda

Interactive Reduct Evolutional Computation for Aesthetic Design

We propose a method to evolve designs based on a user’s personal preferences. The method works through an interaction between the user and a computer system. The objective of the method is to help a customer set design parameters by simple evaluation of displayed samples. An important feature of this method is that the design attributes which a user pays more attention to (favored features) are estimated with Reduct in Rough Sets Theory and are reflected while refining the design. New design candidates are generated by the user’s evaluation of design samples generated at random. While values of attributes estimated as favored features are fixed in the refined samples, the others are generated at random. This interaction continues until the samples converge to a satisfactory design. Thus, this efficient design process evaluates a personal and subjective feature. This method is applied to design a 3D cylinder model such as a cup or a vase. This method is then compared with an Interactive GA.Copyright

Favored Feature Based Shape Generation: Externalization of Latent Preference

Product styling, which is difficult to describe as a specification, has become an important factor in the current saturated market in which many products possess similar functionalities. In this paper, we propose a favored-feature-based design support method which helps the user to externalize his or her image of styling design into a shape without the use of any drawing techniques. The computer system based on this method works through interaction with the user. The user evaluates design samples generated by the system. The system estimates the favored features to which the user unconsciously pays greater attention. The user selects one of the estimated favored features to generate new design samples. The system automatically generates design samples based on the favored feature. The use of the system causes the user to notice his/her own unconscious preferences by displaying showing the favored features. Thus, the user becomes consciously aware of his/her own latent perspectives of design. Through repetition of this operation, the user gradually expands his/her perspectives and can generate a greater variety of design samples based on them. Operation of this system is easy compared with the conventional design system, in which the user must manually adjust the design parameters and which we shall call the manual design system. Therefore the non designer customer can readily express his/her image of style requirements as an actual shape. The proposed method is compared with the manual design system and its effectiveness is demonstrated by a design experiment using a 3D cylinder model consisting of superquadratic ellipse cross sections.Copyright