Controllers for Functional Electrical Stimulation (FES) are still not able to restore natural movements in the paretic arm. In this work, Reinforcement Learning (RL) is used for the first time to control a hybrid upper limb robotic system for stroke rehabilitation in a real environment. The feasibility of the FES controller is tested on one healthy subject during elbow flex-extension in the horizontal plane. Results showed an absolute position error <1.2° for a maximum range of motion of 50°.

An upper limb Functional Electrical Stimulation controller based on Reinforcement Learning: A feasibility case study.

D. Di Febbo;E. Ambrosini;M. Pirotta;M. Restelli;A. Pedrocchi;S. Ferrante
2018

Abstract

Controllers for Functional Electrical Stimulation (FES) are still not able to restore natural movements in the paretic arm. In this work, Reinforcement Learning (RL) is used for the first time to control a hybrid upper limb robotic system for stroke rehabilitation in a real environment. The feasibility of the FES controller is tested on one healthy subject during elbow flex-extension in the horizontal plane. Results showed an absolute position error <1.2° for a maximum range of motion of 50°.
GNB 2018 - SESTO CONGRESSO DEL GRUPPO NAZIONALE DI BIOINGEGNERIA
Functional Electrical Stimulation, Reinforcement Learning, Rehabilitation, Hybrid Robotic Systems.
File in questo prodotto:
File Dimensione Formato  
document_diFebbo_GNB18.pdf

accesso aperto

Descrizione: Articolo principale
Dimensione 230.38 kB
Formato Adobe PDF
230.38 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11311/1070197
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact