An important and growing trend in modern robotics research is to create robots with human-like qualities, which will allow robots to interact naturally with humans and to become be a part of our lives. One of the main goals of socially assistive robotics is to create stimulating and engaging human-machine goal-driven interactions in which the user willingly participates for an extended period of time.

Our work is focused on the interaction modalities, and on the motivational and cognitive strategies for elderly users suffering from cognitive changes related to aging and/or Alzheimer's disease. The main advantages of the proposed approach are that it provides time-extended personalized cognitive and social interaction and "exercise" in a robot-supervised fashion, facilitating ongoing monitoring and companionship. We focus on contact-free strategies, wherein there is no physical contact between the robot and the user, and where social interaction, motivation, and engagement play key roles instead. This is an entirely novel area of research in assistive and rehabilitation robotics and it opens up a broad avenue for future discovery and development.

The specific aims of the proposed study are as follows:

• Validate that a robotic system can safely and effectively interact with elderly users suffering from cognitive changes related to aging and/or Alzheimer's disease.
• Validate that a robotic system can establish a productive interaction with the user that can serve to motivate and remind the user about specific tasks/cognitive exercises.

One basic experimental scenarios will be used:

1. One-on-one interaction between the robot and the user, either in a separate room or in public areas of the senior living facility.

The main scenarios are:

1. Encouraging physical activity: the robot will encourage the user to exercise, providing examples of simple movements and encouraging imitation. All physical activity will be performed slowly, safely, and while providing verbal encouragement.
2. Imitation and control activity: the robot will encourage the user to imitate various arm gestures and poses, providing examples of each, and encourage the user to create novel gestures for the robot to imitate. The user will also be able to use the arm gestures to command the robot to move about the room (e.g. By raising the left arm the robot would move to the left). All physical activity will be performed slowly, safely, and while providing verbal encouragement.

Test-beds

Bandit II, biomimetic anthropomorphic robot

Sample image taken from robot's camera
Person segmentation with key regions identified	Final user arm angle detection output
Sample segmentation for user with dark clothing	Final user arm angle detection output

Here are some videos showing exercise scenario robot behaviors for demonstration purposes: exercise.avi and imitation.avi

• Juan Fasola and Maja J. Matarić. "Robot Motivator: Improving User Performance on a Physical/Mental Task". Poster paper in Proceedings of the International Conference on Human-Robot Interaction (HRI), Mar 2009.[PDF]
  
  
• Adriana Tapus, Juan Fasola, and Maja Mataric' (2008) “Cognitive Assistance and Physical Therapy for Dementia Patients Using Socially Assistive Robots”, In Social Interaction with Intelligent Inddor Robots (SI3R) Workshop, Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), Pasadena, CA, USA, May, 2008. [PDF]
  
  
• Adriana Tapus, Juan Fasola, and Maja Mataric' (2008) “Socially Assistive Robots for Individuals Suffering From Dementia”, In Robotic Helpers: User Interaction, Interfaces and Companions in Assistive and Therapy Robotics Workshop, Proceedings of the 3rd ACM/IEEE International Conference on Human-Robot Interaction (HRI), Amsterdam, The Netherlands, March, 2008. [PDF]
  
  
• Adriana Tapus, Maja J. Mataric´, and Brian Scassellati (2007) "The Grand Challenges in Socially Assistive Robotics", IEEE Robotics and Automation Magazine (RAM), Special Issue on Grand Challenges in Robotics, Vol. 14, No. 1, March, 2007. [PDF]
  
  
• Adriana Tapus and Maja J. Mataric´ (2006) "Towards Socially Assistive Robotics", International Journal of the Robotics Society of Japan (JRSJ), 24(5), pp. 576-578, July, 2006. [PDF]
  
  
• Maja J. Mataric´ and Adriana Tapus (2006) "The Promises and Challenges of Socially Assistive Robotics", 50th Anniversary AI Summit, Monte Verita, Switzerland, July 2006. [PDF]
  
  
• David J. Feil-Seifer and Maja J. Mataric´. "Defining Socially Assistive Robotics". Poster paper in International Conference on Rehabilitation Robotics, pages 465-468, Chicago, IL, Jun 2005. [PDF]
  
  

This work is supported by the USC NIH Aging and Disability Resource Center (ADRC) pilot program and National Academies Keck Futures Initiative (NAFKI) program. We are also grateful to our partners: Silverado Senior Living and Southern California Presbyterian Homes.

Juan Fasola: fasola at usc dot edu