Exercise Support System with Self-Growing Avatars Using Interactive Evolutionary Computation

Abstract

Interactive Evolutionary Computation (IEC) is a technology that optimizes solutions through human-computer interaction based on intuitive human evaluations, and many systems have been proposed. While IEC typically imposes a burden on users due to the repetitive evaluation of solution candidates, there have also been attempts to combine IEC with daily habits, such as dietary routines, to achieve long-term solution optimization. In this paper, we propose a system that links user evaluations in IEC with daily exercise habits. Furthermore, by utilizing the Proteus effect, which visualizes anticipated exercise outcomes, we aim to enhance both exercise motivation and self-efficacy. As a prototype system, we introduce a solution where an avatar evolves toward the user’s ideal body shape through continued exercise using IEC. Experimental results showed that 81.8% of participants experienced increased exercise motivation, confirming improvements in both exercise motivation and self-efficacy.