Getting dressed is a universally performed daily activity, and has a substantial impact on a person’s well-being. Choosing appropriate outfits to wear is important, as clothes protect a person from elements in the environment, and act as a barrier against harsh surfaces [1]. Studies have shown strong correlation between clothing choices and perceptions of sociability, emotional stability, and impression formation (e.g., [2]). This activity, however, can be difficult for some individuals, as they may lack the required reasoning and judgement required [3]. They include children with intellectual and learning disabilities [4] (e.g., Down syndrome [5], dyspraxia [6], autism spectrum disorder [7]), and older adults suffering from dementia including Alzheimer’s disease [8,9], or HIV-associated neurocognitive disorders [10].

In this paper, we present the development of a novel autonomous robotic clothing recommendation system to provide appropriate clothing options, which are personalized to a user’s wardrobe. This research expands on our previous work on socially assistive robots providing assistance with other daily activities, including meal eating [11] and playing Bingo games [12]. Currently, a few smartphone applications exist for providing outfit choices (e.g., [13,14]); however, unlike our proposed system, they are fashion-focused and not able to adapt online to a user’s preferences. Furthermore, by utilizing a socially assistive robot, we provide a more engaging interaction.

We utilize the small Nao social robot, Leia, to guide and interact with a user in order to obtain information regarding his/her preferences, the activity for which the clothing will be worn, as well as the environment in which the activity will take place in order to make outfit recommendations, Fig. 1.

This content is only available via PDF.