University of Twente Student Theses


Synthetic Faces For Robots

Bosschart, J.S.P. (2023) Synthetic Faces For Robots.

[img] PDF
Abstract:In the field of Human-Robot Interaction (HRI), the appearance of robots plays an important role as it influences users’ perceptions and expectations of their behaviour, shaping the overall user experience. This research project explores the potential of generative artificial intelligence (AI) as a design tool to assist robot designers in making informed decisions about robot face design. The central research question addressed in this study is, “How to create a tool that simplifies image generation to generate robot faces to assist robot designers in their decision-making?” To answer this question, the project examines ways to integrate AI-driven image generation models into robot face design to enhance the overall design process. Additionally, the project investigates the potential value of these tools in educational and design contexts. The research methodology comprises a background research phase, including a literature review on essential factors in designing robot faces, as well as an examination of the latest advancements in digital decision-making and AI tools, gathering new insights for incorporating into a novel tool. The research further utilises the creative technology design method. As part of the ideation phase, a stakeholder analysis was conducted to understand the perspectives and needs of users. The analysis, together with the background research, served as a foundation for brainstorming and ideation, of which two concepts were chosen for further development into prototypes. The first prototype is an interactive software solution that guides users through setting up their own image generation model, with a specific focus on robot face-related imagery. This solution is implemented as a Jupyter Notebook, offering the versatility to run the file on cloud servers, given the resource-intensive nature of image generation. The second prototype is an informative guide presented as a poster, providing insights and guidelines for AI-driven robot face design. This guide is tailored to beginners in the AI field and is compatible with various existing image-generation tools, enabling easy sharing on social media platforms for increased accessibility. While the prototypes show potential, further user testing is required to validate their effectiveness in enhancing the robot face design process. The study aims to contribute to the advancement of Human-Robot Interaction and AI-driven design tools, facilitating more efficient and creative decision-making processes in the field of robotics.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:01 general works
Programme:Creative Technology BSc (50447)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page