Robot-Guided Exercise Program for the Rehabilitation of Older Nursing Home Residents : Page 2 of 2

June 17, 2013

Study Limitations

Our study has several limitations. First, our study population was small, as we tested the robot in small nursing homes, which provided a more intimate, home-like environment; thus, whether the robot would be feasible in a larger, more institutional nursing home setting is unclear. Second, the robot itself was small, which limited the types of exercises we could program into the robot; for example, finger manipulation exercises would be difficult for the robot to demonstrate, particularly in a group setting. Third, we did not evaluate the robot’s impact on the physical health and well being of residents, so it is not clear whether a robot can serve as a suitable substitute to a physical therapist and/or a fitness trainer with regard to leading exercise sessions that do not require monitoring by such professionals.


Our current study was not the first time we used an NAO robot in the nursing home setting. Previously, we successfully used this robot as part of a remote monitoring system that was developed for nursing homes.4 The robot received alarms that were either automatic or activated by the residents via its wireless Internet connection and navigated independently to the room where the alarm originated. After it entered the room, the robot transmitted near–real-time images to the staff and opened a voice connection between the resident and the remote caregiver(s). This enabled the remote caregiver(s) to check on the patient and take appropriate action if intervention was required. The nursing home residents and staff responded positively to this remote monitoring system.4

Because physical activity is essential for elderly nursing home residents, particularly those who are sedentary, we decided to evaluate the NAO robot as an exercise trainer. Conventional rehabilitation generally involves one-on-one interaction with a physical therapist and/or fitness trainer, both of who assist and encourage the patient through repetitive exercises5; however, one-on-one rehabilitation is not always an option due to factors such as financial limitations. In these situations, interactive robotic therapists can provide reproducible exercise programs (ie, exercise activities that residents can perform without the supervising presence of a therapist), and these robotic therapists can be programmed to carry out a number of exercise programs of different levels and requirements. These programs can be individually tailored based on patients’ physical and cognitive abilities. Because different programs can be preprogrammed into the robot’s memory, there would be no need for special staff to take care of the robot. In addition, the same robot could be used for multiple purposes, such as for monitoring activities and as an exercise trainer.

Currently, there are two basic types of robots that can be used in healthcare settings: assistive robots and socially interactive robots.6 Assistive robots provide assistance to the user, such as by performing a physical task that the patient or caregiver is unable to perform. When we used the NAO robot as a part of a remote monitoring system, it functioned as an assistive robot, alleviating the staff burden of having to physically check on every alarm. Socially interactive robots communicate with the user through social and nonphysical interaction. We used the NAO robot in this manner for the robot-guided exercise programs, as it led residents through an exercise program and provided them with prompts to facilitate interactivity. In most cases, assistive robots capable of physical interaction are the best choice for facilitating geriatric care.

In the nursing home, robots have been used for a variety of purposes. For instance, they have been used to support nursing home staff with the residents’ daily activities7; typical examples of the tasks carried out by robots include improving drug compliance,8 increasing physical exercise adherence,9 playing the role of a companion animal,10 or assisting in different kinds of rehabilitation tasks, such as those targeted to stroke survivors.5 Additionally, Cesta and colleagues11 described a monitoring approach in which a robot worked as part of an intelligent home environment as a means of enabling elderly persons to age in place.

Although the ultimate goal of robotics may be to develop humanoid robots that independently perform a vast gamut of functions, rather than single tasks, robotics is still in its infancy. Currently, humanoid robots are clumsy and expensive, making them a poor substitute for human caretakers. To address this problem, Japan, which has a growing elderly population and a critical shortage of caretakers, launched a “Home-use Robot Practical Application Project” in 2009, after it had spent billions of yen focused on humanoid robotics research.12,13 Through this project, the Japanese government has increased its focus on commercializing simple robots that perform single, focused tasks as a means of enabling more elderly persons to age in place. In addition, the Japanese government has announced that starting in 2013, it is providing subsidies to firms working on nursing care robots.12 These subsidies are projected to cover 50% to 66% of research and development costs. Four specific types of assistive nursing care robots are included in this plan: (1) a motorized robot suit that can facilitate caretakers lifting and moving physically impaired patients; (2) an ambulatory robot that enables elderly persons and others with impaired mobility to walk by themselves, including on inclines; (3) a monitoring robot that can track the movement of dementia patients and indicate their location; and (4) a portable, self-cleaning robot toilet that can facilitate toileting activities.12 The government’s goal is to alleviate the chronic shortage of nursing care workers by promoting a variety of low-cost nursing care robots.

The humanoid robot we used in our research was not specifically designed to be an exercise trainer or a remote monitoring system, but was successfully programmed to carry out these functions. Our findings indicate that until humanoid robots become more sophisticated, the current generation of humanoid robots can still serve important functions in the nursing home setting; however, at a cost of approximately $16,000 each, these robots could remain cost prohibitive for some facilities. Nevertheless, as robotics technology continues to advance, it will become increasingly more affordable. We also speculate that once humanoid robots become affordable and capable of performing a vast variety of tasks independently, there will still be a need for a wide variety of simple, single-function robots.


The feedback we received regarding use of the NAO robot as an exercise trainer has been mainly positive, indicating that use of a humanoid robot for the rehabilitation of elderly nursing home residents is feasible and warrants further investigation. One important component of future research is to examine the long-term health outcomes of nursing home residents who engage in robot-guided exercise programs versus those who undergo such programs with human therapists. In addition, use of the robot to engage residents in other social activities should be explored, as the exercise activities we programmed into our NAO robot serve as only one example of how these robots can be used to socially interact with and engage residents. With all of the enablers required for independent movement and interaction with people, the NAO robot could be used to guide residents through a wide variety of mentally and physically stimulating activities, opening up new areas of development and research.


1.     Pollack ME. Intelligent technology for an aging population: the use of AI to assist elders with cognitive impairment. AI Magazine. 2005;26(2).

2.     Liu PR, Meng MQ, Liu PX, Tong FF, Chen XJ. A telemedicine system for remote health and activity monitoring for the elderly. Telemed J E Health. 2006;12(6):622-631.

3.     Buman MP, Hekler EB, Haskell WL, et al. Objective light-intensity physical activity associations with rated health in older adults. Am J Epidemiol. 2010;172(10):1155-1165.

4.     Bäck I, Kallio J, Perälä S, Mäkelä K. Remote monitoring of nursing home residents using a humanoid robot. J Telemed Telecare. 2012;18(6):357-361.

5.     Krebs HI, Volpe BT, Williams D, et al. Robot-aided neurorehabilitation: a robot for wrist rehabilitation. IEEE Trans Neural Syst Rehabil Eng. 2007;15(3):327-335.

6.     Feil-Seifer D, Mataric MJ. Socially assistive robotics. IEEE Robotics and Automation Magazine. 2011;18(1):24-31.

7.     Pineau J, Montemerlo M, Pollack M, Roy N, Thrun S. Towards robotic assistants in nursing homes: challenges and results. Robotics and Autonomous Systems. 2003;42(3-4):271-281.

8.     Takacs B, Hanak D. A prototype home robot with an ambient facial interface to improve drug compliance. J Telemed Telecare. 2008;14(7):393-395.

9.     Gadde P, Kharrazi H, Patel H, MacDorman KF. Toward monitoring and increasing exercise adherence in older adults by robotic intervention: a proof of concept study. Journal of Robotics. Published online 2011. Accessed May 30, 2013.

10.   Tamura T, Yonemitsu S, Itoh A, et al. Is an entertainment robot useful in the care of elderly people with severe dementia? J Gerontol A Biol Sci Med Sci. 2004;59(1):M83-M85.

11.   Cesta A, Cortellessa G, Rasconi R, Pecora F, Scopelliti M, Tiberio L. Monitoring elderly people with the robocare domestic environment: interaction synthesis and user evaluation. Computational Intelligence. 2011;27(1):60-82.

12.   Shimbun Y. Japan to promote robots for nursing home care. BDN Maine. http://ban
. Published April 29, 2013. Accessed June 1, 2013.

13.   Hyashi T. Japan promotes practical application of home-use robots. Tech On! Published August 5, 2009. Accessed June 1, 2013.


Disclosures: The authors report no relevant financial relationships.

Address correspondence to: Iivari Bäck, PhD, University of Vaasa, Department of Management

P.O. Box 700, FI-65101 Vaasa, Finland;