革新知能統合研究センター

Cognitive Behavioral Assistive Technology (CBAT) Team, RIKEN Center for Advanced Intelligence Project (CBAT-RIKEN-AIP) puts emphasis on developing CBAT which promotes cognitive reserve for prevention of cognitive decline and dementia of older adults which damage the intelligence of human necessary for social life.
Special Interest Group on “Nursing Care Robot” of the Robotics Society of Japan (SIGNCR-RSJ) was established to study engineering, humanities and sociological considerations of robots that are penetrating the nursing care field and the systematization of “nursing care robotics”.
We invite Rayna Hata who develops assistive technology for preventive care of older adults.
This seminar is co-hosted by CBAT, RIKEN AIP (CBAT-RIKEN-AIP) and Special Interest Group on “Nursing Care Robot” of the Robotics Society of Japan (SIGNCR-RSJ).

Date and Time: July 16, 2025: 10:30 – 11:30 (JST)
Venue: Online and Open Space at the RIKEN AIP Nihonbashi office
*Open Space is available to AIP researchers only

Host: Cognitive Behavioral Assistive Technology Team, RIKEN Center for Advanced Intelligence Project (CBAT-RIKEN-AIP)
Special Interest Group on “Nursing Care Robot” of the Robotics Society of Japan (SIGNCR-RSJ)

Program
10:30-11:30:
Speaker: Rayna Hata, CMU
Title: Robotic Exercise Buddy: Motivating Older Adults to Exercise through Social and Personalized Coaching
Abstract:
Exercise adherence among older adults is critical for healthy aging, yet motivation to exercise often declines with age. At the same time, the rapidly growing aging population places increasing strain on caregivers, physical therapists, and exercise coaches, making it increasingly difficult to provide consistent, individualized support. To address these challenges, we present a robotic exercise system designed to enhance motivation and adherence through two distinct interaction styles: a coach-style robot that delivers structured, performance-oriented feedback, and a social buddy-style robot that engages users in natural conversation and companionship during exercise. The coach-style robot uses a contextual bandit algorithm to adaptively switch between firm and encouraging one-way feedback styles based on the user’s daily preferences and performance. In contrast, the social buddy leverages a large language model (LLM) system to generate contextually relevant, human-like dialogue that builds rapport and fosters engagement. Through a user study with older adults, we examine individual preferences for coaching style, develop a personalized and adaptive approach to robot behavior, and introduce methods for measuring motivation and conversational engagement in real time. In this talk, we present findings from the study and share future directions for designing socially intelligent, adaptive robotic systems that promote health and well-being in aging populations.

Rayna Hata Profile:
Rayna Hata is a Ph.D. student at the Robotics Institute at Carnegie Mellon University, where she is a member of the Transportation, Bots, and Disability Lab, advised by Dr. Aaron Steinfeld. Her research interests are Human-Robot Interaction (HRI) and assistive technology, with a particular focus on how robots can engage people in meaningful, personalized ways. She explores how conversational dynamics and individual user preferences influence engagement, trust, and long-term use of robotic systems. Her work combines real-time sensor data, adaptive robot behavior, and large language models to create more effective and natural interactions between humans and robots.
She is currently interning at the Assistive Technology Lab at the National Museum of Emerging Science and Innovation (Miraikan) in Tokyo, Japan, where she is conducting research with the “AI Suitcase”, an autonomous, navigation robot that guides visually impaired people.