Generating Shared Latent Variables for Robots to Imitate Human Movements and Understand their Physical Limitations

Assistive robotics and particularly robot coaches may be very helpful for rehabilitation healthcare. In this context, we propose a method based on Gaussian Process Latent Variable Model (GP-LVM) to transfer knowledge between a physiotherapist, a robot coach and a patient. Our model is able to map visual human body features to robot data in order to facilitate the robot learning and imitation. In addition, we propose to extend the model to adapt the robots’ understanding to patients’ physical limitations during assessment of rehabilitation exercises. Experimental evaluation demonstrates promising results for both robot imitation and model adaptation according to patients’ limitations.

[1]  J. Buckwalter,et al.  The burden of musculoskeletal conditions at the start of the new millennium , 2003 .

[2]  Franz Kummert,et al.  Exercising with a humanoid companion is more effective than exercising alone , 2016, 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids).

[3]  Albert Ali Salah,et al.  A Robotic Fitness Coach for the Elderly , 2013, AmI.

[4]  Maxime Devanne,et al.  A Co-design Approach for a Rehabilitation Robot Coach for Physical Rehabilitation Based on the Error Classification of Motion Errors , 2018, 2018 Second IEEE International Conference on Robotic Computing (IRC).

[5]  Rajesh P. N. Rao,et al.  Learning Shared Latent Structure for Image Synthesis and Robotic Imitation , 2005, NIPS.

[6]  Behzad Dariush,et al.  Online Transfer of Human Motion to Humanoids , 2009, Int. J. Humanoid Robotics.

[7]  Martin Fodslette Møller,et al.  A scaled conjugate gradient algorithm for fast supervised learning , 1993, Neural Networks.

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

[9]  Neil D. Lawrence,et al.  Gaussian Process Latent Variable Models for Visualisation of High Dimensional Data , 2003, NIPS.

[10]  Stefan Kopp,et al.  Realizing a low-latency virtual reality environment for motor learning , 2015, VRST.

[11]  Maren Bennewitz,et al.  Real-time imitation of human whole-body motions by humanoids , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[12]  Matthieu Lapeyre.  Poppy : open-source, 3D printed and fully-modular robotic platform for science, art and education. (Poppy : plate-forme robotique open source, imprimée en 3D et totalement modulaire pour l'experimentation scientifique, artistique et pédagogique) , 2014 .

[13]  P. Kent,et al.  The efficacy of targeted interventions for modifiable psychosocial risk factors of persistent nonspecific low back pain - a systematic review. , 2012, Manual therapy.

[14]  Catherine Pelachaud,et al.  The TARDIS Framework: Intelligent Virtual Agents for Social Coaching in Job Interviews , 2013, Advances in Computer Entertainment.

[15]  Albert Ali Salah,et al.  An autonomous robotic exercise tutor for elderly people , 2017, Auton. Robots.

[16]  Maxime Devanne,et al.  Multi-level motion analysis for physical exercises assessment in kinaesthetic rehabilitation , 2017, 2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids).

[17]  Ales Ude,et al.  Enabling real-time full-body imitation: a natural way of transferring human movement to humanoids , 2003, 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422).

[18]  Juan Fasola,et al.  A socially assistive robot exercise coach for the elderly , 2013, J. Hum. Robot Interact..

[19]  C. Stanton,et al.  Teleoperation of a humanoid robot using full-body motion capture , example movements , and machine learning , 2012 .

[20]  Takenori Obo,et al.  Imitation learning for daily exercise support with robot partner , 2015, 2015 24th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN).

[21]  Joaquin Quiñonero Candela,et al.  Local distance preservation in the GP-LVM through back constraints , 2006, ICML.