Zhidong Su is a Ph.D. candidate in the School of Electrical and Computer Engineering at Oklahoma State University, advised by Prof. Weihua Sheng. His research focuses on elderly care companion robot, conversational AI, learning and adaptation.
As people age, they often face various physical or cognitive challenges, therefore requiring assistance from their families or caregivers, which is becoming a significant burden for society. Companion robots are equipped with many functions to assist older adults with their daily lives. However, how to make companion robots user-friendly for older adults and adapt to their specific needs or characteristics is still an open question. My current research aims to improve the user-friendliness and intelligence of companion robots by designing new learning and adaptation mechanisms, which are utilized in a medication management system and a proactive context-aware conversation system. I proposed reinforcement learning-based and large language model-based adaptation methods that enable a companion robot to learn from interactions with users and adapt to their preferences with minimal feedback, enhancing its practicality for real-world usage.