An intelligent socially assistive robot-wearable sensors system for personalized user dressing assistance

Individuals living with cognitive impairments are faced with unique challenges in completing important activities of daily living such as dressing. In this paper, we present the first socially assistive robot-wearable sensors system to provide dressing assistance through social human-robot interactions. A novel robot-wearable architecture has been developed to recognize and classify user dressing actions and provide personalized prompts and feedback. Our system uses smart clothing with embedded strain sensors to estimate the user's actions, which are then classified into different dressing steps. Our assistive robot uses a MAXQ hierarchical learning method to learn appropriate assistive behaviors to aid a user with the sequence of dressing steps. Experiments conducted validated the performance of our robot-wearable system in identifying and effectively responding to a variety of user states and dressing step actions. Furthermore, a robot demonstration study with stakeholders found that overall, they had positive perceptions and attitudes towards the socially assistive robot-wearable system, in particular with respect to its usefulness with the intended user population.