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

TL;DR

This paper introduces a GP-LVM based method enabling robots to imitate human movements and adapt to patients' physical limitations, improving assistive rehabilitation through shared latent representations.

Contribution

It presents a novel GP-LVM approach for transferring human movement knowledge to robots and adapting to individual patient limitations in rehabilitation tasks.

Findings

Successful robot imitation of human movements

Effective adaptation to patients' physical limitations

Promising results in rehabilitation scenarios

Abstract

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 robots' understanding to patient's physical limitations during the assessment of rehabilitation exercises. Experimental evaluation demonstrates promising results for both robot imitation and model adaptation according to the patients' limitations.