Ryan: A Companion Robot for Seniors
A team of engineers at the University of Denver led by Prof. Mohammad Mahoor, is developing a socially-assistive robot called Ryan that provides companionship to elderly people with dementia and/or depression. Dementia is an overall term for diseases that deteriorate individuals’ memory and other mental skills. Dementia can significantly reduce elderly individuals’ ability to live independently and safely in their homes. Associated with the decline in cognitive abilities, depression is often one of the symptoms of dementia. Due to dementia and the rapid aging of the population, nursing homes have been facing a challenge to provide care. There is thus a critical and growing demand in the community to find effective ways to provide care for elderly people with dementia/depression. There is an emerging research field in robotics that aims to use social robots to engage effectively in social and conversational interaction with seniors with dementia to improve their socio-emotional behaviors, cognitive functions, and well-being.
We at DreamFace Technologies, LLC, a start-up company co-founded by Dr. Mahoor, partnered with the University of Denver and the Knoebel Institute for Healthy Aging (KIHA) work to develop Ryan, for this purpose. Ryan is a lifelike social robot with the capability of showing facial expressions, visual speech, emotion recognition, subject movement tracking, and eye gaze. She can maintain a spoken dialog and is designed for face-to-face communication with individuals in different social and therapeutic contexts.
One of the key features of the robot is its animated face. The face uses a patented rear-projection system that allows her to change her appearance to appeal to the users and be more expressive. Also, the animated face helps easily sync her “lips” to her voice, which is important for understanding her, especially for elderly individuals.
Another interesting feature of Ryan is her ability to recognize a user’s emotions through Artificial Intelligence. This ability allows her to understand the user’s mood and react appropriately. Combining this ability with her expressive face gives her a sense of empathy and helps create a stronger bond with the user.
Ryan is not just a pretty face though, users can have meaningful conversations with her too. Unlike Amazon Alexa or Google Home, she is not an assistant, but a companion. Ryan is proactive, she does not only answer your questions, she can start a conversation about anything and keep the user engaged.
The robot is also equipped with a screen on its torso with features such as cognitive games, a music player, narrated photo albums, and a video player. It even reminds the users to take their medicine on time and stay on schedule. The simple and interactive cognitive games are based on Montessori activities which have been developed to help slow down the cognitive decline associated with dementia.
We conducted a pilot study to demonstrate the feasibility of using Ryan to improve the quality of life of seniors with moderate dementia and/or depression. So far Ryan has been tested with 12 seniors. Living with them for an average of four weeks, she provided constant companionship and mental stimulation through cognitive games and conversation. Overall, the seniors felt the robot helped them maintain their schedule, improved their mood, and stimulated them mentally. The common sentiment among users after the pilot study was best described by one user’s comment, “She [Ryan] was just enjoyable. We were SAD to see her go.” Looking toward the future, the second version of Ryan is already being developed with new features to further improve seniors’ quality of life. The company has received fast-track SBIR grants from NSF and NIH/NIA.
Dr. Mahoor’s bio:
Dr. Mohammad H. Mahoor is a Professor of Electrical and Computer Engineering at DU. He does research in the area of computer vision and machine learning including visual object recognition, object tracking and pose estimation, motion estimation, 3D reconstruction, and human-robot interaction (HRI) such as humanoid social robots for interaction and intervention with children with special needs (e.g. autism) and elderly with depression and dementia. He has received over $6M of research funding from state and federal agencies including the National Science Foundation and National Institute of Health. He is a Senior Member of IEEE and has published more than 120 peer-reviewed scientific conference and journal papers. Dr. Mahoor received his Ph.D. in Electrical and Computer Engineering from the University of Miami, Florida in 2007.