PHYSICAL ROBOT  TRAINER

RESEARCH SUMMARY

A robotic system is developed as a personal coach for older adults. The system aims to motivate older adults to participate in physical activities. The robot instructs the participants and demonstrates the exercises. It also provides real-time corrective and positive feedback according to the performance of the participants as monitored by an RGB-D Kinect camera. Two robotic systems were developed and implemented using the Python programming language.
Experimental studies, composed of a preliminary experiment followed by an experiment with two different humanoid robots (Nao and Poppy), aimed to determine the best timing and mode of feedback that would accommodate user preferences, motivate the users and improve their interaction with the system. A comparative study was also carried out to explore user preferences while training with the two forms of humanoid robots and to determine which of the robots gives better satisfaction.
A preliminary experiment was conducted with ten older adults, aged 67-85, in a home-like environment to keep the interaction as natural as possible. A simulated form of the Nao robot was used, due to a failure in the real robot. Yet, the results revealed that 70% of the participants expressed their intention to use the system in the future.
This gave the motivation for a follow-up study with thirty-two older participants, aged 70-88. The experiment was a between-within design where the independent variables were the timing and mode of feedback and the type of humanoid robot. Each older adult interacts with Nao and Poppy robots at different sessions in a randomized order as the feedback preferences are being assessed. The dependents variables are perceived usefulness, ease of use, attitude and behavioral intention to use which were assessed both objectively and subjectively. The study reveals the potential of a robotic system to provide older adults with the independence and motivation to participate more effectively in physical activities which is beneficial to their health and general wellbeing and provides specific design guidelines.
The results revealed that the system fulfills the aim of motivating older adults to engage more in physical exercises. Most of the users perceived the system as very useful and easy to use. Users had a positive attitude towards the system and noted their the intention to use it. Continuous feedback significantly increased positive attitude, engagement and ease of use of the system. The results also revealed that the Poppy robot engaged the users more than the Nao robot. Audio and visual feedback is the preferred mode of feedback with regards to ease of use of the system and the engagement.