User Studies with Intelligent Assistive Robots and Elder Residents Living in Long-Term Care Homes

Publications

Tran, T., Vaquero, T., Beck, J.C., Nejat, G. (May 2015). Robots in Retirement Homes: Applying Off-the-Shelf Planning and Scheduling to a Team of Assistive Robots. To be submitted to Journal of Artificial Intelligence Research.

Louie, W.G., Li, J., Mohamed, C., Despond, F., Lee, V., Nejat, G. (April 2015). Tangy the Robot Bingo Facilitator: A Performance Review. Journal of Medical Devices-Transactions of the ASME.

Vaquero, T., Mohamed, C., Nejat, G., Beck, J.C. (2015). The Implementation of a Planning and Scheduling Architecture for Multiple Robots Assisting Multiple Users in a Retirement Home Setting. AAAI ’15 Workshop on Artificial Intelligence Applied to Assistive Technologies and Smart Environments, pp. 1-6.

Louie, W.G., Li, J., Vaquero, T., Nejat, G. (2015). Design Considerations and Impressions of Socially Assistive Robots for Seniors Living in Long-term Care Facilities. Human-Robot Interactions: Principles, Technologies, and Challenges. Nova Publishers.

Vaquero, T. (2014). A Planning & Scheduling Module for a Multi-Robot and Multi-User Centralized System Architecture. Internal Report. Department of Mechanical and Industrial Engineering. University of Toronto.

Mohamed, C., Vaquero, T., Nejat, G. (2014). Tangy the Socially Assistive Robot: Stimulating Recreational Activities at Long-term Care Facilities. Submitted to the Canadian Geriatrics Journal.

Li, J., Louie, W.G., Vaquero, T., Nejat, G. (2014). Socially Assistive Robots: A Focus Group Study. Submitted to the Canadian Geriatrics Journal.

Rationale: Assistive robots are seen as a promising new technology to address the needs of our aging population. However, there are only a handful of research studies that investigate assistive robot interactions with elderly individuals in real healthcare environments. The robotics aspects of developing the technology are challenging (e.g. robots have to autonomously find people, navigate and localize themselves in a dynamic human-centred environment). However, the human-robot interaction is equally or more crucial for integrating these robots as effective long-term resources: robots must determine their assistive behaviours based on information from residents and must adapt their behaviours in order to be accepted and used by the residents. Without direct user studies that investigate these challenges, there is little likelihood that such robot technology can be successfully deployed.

Objective: The objective of this one-year project was to conduct onsite user experiments with an intelligent socially assistive robot that was capable of assisting elders in multiple socially and cognitively stimulating activities, both one-on-one and in a group setting.

Research Plan: We conducted extensive studies in a long-term care (LTC) facility (O’Neill Centre in Toronto) with our autonomous assistive robot and elderly residents, to investigate engagement, compliance, acceptance, and performance of the robot for both video chats with family members/friends and a multi-person game of Bingo.

Lay Title: Smart and helpful robots welcomed by older adults living in long term care

Key Findings:

• Tangy was able to effectively plan and schedule the daily activities of the robot to facilitate telepresence sessions and Bingo games while considering the requests and schedules of users. The robot was also able to sense its environment and the players during interactive Bingo games, while appropriately determining its assistive behaviours.
• A 99.23% compliance rate was determined for the Bingo interaction with Tangy.
• 90% of the total interaction time, players had visual focus of attention either towards Tangy or their Bingo cards. The remaining time they were looking at and talking with other players and looking at these players’ cards, as well as looking around the room. It is important to note that visual focus of attention towards the robot increased when Tangy was providing personalized assistance.
• Residents were comfortable and enjoyed interacting with the robot and would interact with it again in the future.

Why was this study needed?

There is an increasing need to provide effective services, care, and companionship to our aging population. Assistive robots can provide a potential solution for a decreasing workforce and an increasing senior population. However, little research activity exists in Canada on the development and use of robotics for eldercare. Namely, there is an urgent need to promote serious initiatives to nurture healthcare robotics research and their subsequent commercialization to improve the quality of life of one of our most vulnerable populations.

Brief overview of the methodology:

• We developed and implemented a novel and needed assistive robotic technology to provide social and cognitive interventions to the elderly in LTC facilities in order to improve their quality of life and health outcomes.
• The objective was to design and implement the socially assistive robot Tangy capable of autonomously facilitating socially and cognitively stimulating activities for elderly individuals living in LTC facilities.
• For this exploratory study, Tangy was placed at the O’Neill Centre to investigate if it can effectively engage elderly residents in multi-user activities. Specifically, human-robot interaction sessions were conducted with Tangy as a Bingo facilitator. We investigated: system performance, engagement and compliance, the acceptance and attitudes of residents towards Tangy.
• We included individuals aged over 60 years, who were fluent in English, could hear normal levels of speech; and were cognitively intact or had mild cognitive impairment.

Potential impact of findings on clinical practice/patient care and how this impact might be measured:

• The research program has focused on addressing the social challenges we are facing with an aging population and the role that robots can play in assisting our elderly to promote quality of life through socially and cognitively stimulating recreational activities.
• Our research investigated the validity of using these robots in LTC to benefit this population and furthered discussions on the use of such robots with our elderly population.
• Through outreach to the LTC community based on our study, we have had significant interest from other LTC facilities for us to pilot the robot there to obtain additional results from multiple sites. This would be important in analyzing the needs of different sites and ensuring that the robot can be used across multiple facilities and with an increased number of users.
• Going forward we have created a dialogue with healthcare institutions regarding best practices with such technologies as well as social and economic impacts.

Remaining knowledge/research gaps:

• The user study with Tangy is the first of its kind in Canada. This research will help solidify Canada’s role and presence on the international level in the development of assistive robotics for the elderly to meet the challenges we as Canadians and the rest of the world are facing due to changing demographics.
• We need to continue working closely with clinicians to develop and incorporate such robotic technology for effective long-term implementation and use in LTC facilities.