More Sensors, More Assistive Functions? A Scoping Review of Multimodal Sensing and Assistive Functions in Socially Assistive Robots for Older Adults

Abstract

Socially Assistive Robots (SARs) have gained increasing relevance in older adult care due to their potential to provide companionship, social interaction, and support for activities that promote autonomy. However, barriers to their adoption remain, as older adults particularly value functions that provide practical and meaningful support in their daily lives. In this context, multimodal sensing can expand the range of information available to the robotic system, although the extent to which these additional inputs contribute to assistive functions remains unclear. This scoping review analyzes the literature published over a ten-year period on multimodal SARs designed for older adults, with particular emphasis on the sensors and devices used, multimodal integration strategies, and their relationship with the implemented assistive functions. A total of 33 studies were included, identifying an average of 3.27 sensor or device categories and 3.03 assistive functions per study. Audio and vision were the predominant modalities, while social facilitation, companionship, and emotional support were the most frequently reported functions. In addition, multimodal integration or fusion was explicitly reported in only a limited number of studies. The findings highlight the need to move toward SARs that, while maintaining their social capabilities, more clearly integrate multimodal information with practical assistive functions aimed at supporting older adults’ autonomy and daily activities.