Nature-Inspired Design Idea Generation With Generative Adversarial Networks

Generating new, creative, and innovative ideas in the early stages of the design process is crucial for developing better and original products. Human designers may become too attached to specific design ideas, preventing them from generating new concepts and achieving ideal designs. To come up with original design ideas, a designer needs to have a creative mind, as well as knowledge, experience, and talent. Verbal, written, and visual sources of inspiration can also be valuable for generating ideas and concepts. This study presents a visual integration model that uses a data-supported Artificial Intelligence (AI) method to generate creative design ideas. The proposed model is based on a generative adversarial network (GAN) that combines target object and biological object images to produce new creative product images inspired by nature. The model was successfully applied to an aircraft design problem and the resulting sketches inspired designers to generate new and creative design ideas and variants in a case study. It was seen that this approach improved the quality of the ideas produced and simplified the idea and concept generation process.