Auto-evaluation of motion imitation in a child-robot imitation game for upper arm rehabilitation

The purpose of this study is fusing play-like child robot interaction with physiotherapy in order to achieve upper arm rehabilitation by motivating the child. The proposed system is not intended to substitute for the physiotherapist, but to assist them in their therapeutic tasks by encouraging the child's participation in the activity. Recognizing the imitation performance of the child and supporting him/her with feedback for drawing the child's attention and motivating the child to imitate the robot is crucial. This study concentrates on automatically evaluating the upper body actions of the child during an imitation based physical therapy. For quantifying the performance of the child, two measures were considered: Range of Motion (RoM) and Dynamic Time Warping (DTW) distance. In our initial experiments, eight healthy children were asked to stand in front of a Kinect sensor and to mimic the actions of the humanoid robot Nao, which consist of shoulder abduction, shoulder vertical flexion&extension and elbow flexion. The proposed evaluation measure is verified as a reliable measurement according to Intraclass Correlation Coefficient (ICC) through comparison with evaluations of five physiotherapists as ground truth. The degree of consistency among our ratings and the physiotherapist ratings is between %76 and %96 for different motions.

[1]  Fabio Tesser,et al.  Multimodal child-robot interaction: building social bonds , 2013, HRI 2013.

[2]  Kerstin Dautenhahn,et al.  A Pilot Study with a Novel Setup for Collaborative Play of the Humanoid Robot KASPAR with Children with Autism , 2014, Int. J. Soc. Robotics.

[3]  Juan Manuel Ibarra Zannatha,et al.  Development of a system based on 3D vision, interactive virtual environments, ergonometric signals and a humanoid for stroke rehabilitation , 2013, Comput. Methods Programs Biomed..

[4]  Ayanna M. Howard,et al.  Quantifying upper-arm rehabilitation metrics for children through interaction with a humanoid Robot , 2012, HRI 2012.

[5]  Gerhard Tröster,et al.  Monitoring motor capacity changes of children during rehabilitation using body-worn sensors , 2013, Journal of NeuroEngineering and Rehabilitation.

[6]  Dan O. Popa,et al.  Human-Robot Upper Body Gesture Imitation Analysis for Autism Spectrum Disorders , 2013, ICSR.

[7]  Marina Fridin,et al.  Robotics Agent Coacher for CP motor Function (RAC CP Fun) , 2014, Robotica.

[8]  D. Feil-Seifer,et al.  Defining socially assistive robotics , 2005, 9th International Conference on Rehabilitation Robotics, 2005. ICORR 2005..

[9]  Roderic A. Grupen,et al.  Multidisciplinary stroke rehabilitation delivered by a humanoid robot: Interaction between speech and physical therapies , 2013 .

[10]  G. Landreth Play Therapy: The Art of the Relationship , 1991 .

[11]  Meinard Müller,et al.  Dynamic Time Warping , 2008 .

[12]  D. J. White,et al.  Measurement Of Joint Motion A Guide To Goniometry , 2016 .

[13]  J. Plewes,et al.  Measurement of joint motion: A guide to goniometry , 1986 .