Page 344 - Handbook of Biomechatronics
P. 344
Upper Extremity Rehabilitation Robots: A Survey 337
In the human body, the arm motion is controlled by the CNS, so control-
lers that have any characteristics of the CNS might be advantageous for reha-
bilitation robotics. Since triggered passive controllers are dealing with
biosignals, they can provide powerful tools for rehabilitation by inclusion
of biological feedback. Thus, recent developments in rehabilitation robotics
aremostlyfocusedonleveragingthesetypeofcontrollerstoimprovethequal-
ity of biologically plausible therapy. Furthermore, model-based controllers
(e.g.,NMPC)canalsoprovidebiomechanicallyplausibletools forrehabilitation;
consequently, some studies in recent years have been focused on this idea.
Traditional physical therapies suffer from various inadequacies
(Jorgensen et al., 1995; Ifejika-Jones and Barrett, 2011) and may result in
significant financial burdens from costly therapy sessions (Dong et al.,
2006; Krebs and Hogan, 2012). It is important to continue advancing reha-
bilitation robots, supported by innovative motor learning scenarios (Brewer
et al., 2007; Cano-de-la Cuerda et al., 2015) and the optimization of
mechatronic design and control algorithms, since they can result in effective
in-home rehabilitation and patient care (Dong et al., 2006; Poli et al., 2013).
Furthermore, these interactive and friendly robots can provide variations in
delivering therapy (building on new achievements in motor learning studies)
(Brewer et al., 2007; Reinkensmeyer, 2009), and meaningful restoration of
functional activities (Krebs and Volpe, 2013). In conclusion, we fully expect
that more progress will be made in the near future to improve the design
and control of rehabilitation robots for providing biologically plausible
autonomous therapy.
GLOSSARY
Artificial intelligence
AI
Brain-computer interface
BCI
Central nervous system
CNS
Degree-of-freedom
DOF
Electroencephalogram
EEG
Electromyography
EMG
Functional electrical stimulation
FES
Functional magnetic resonance imaging
fMRI
Functional near-infrared spectroscopy
fNIRS
Human-robot interaction
HRI
Model-based system engineering
MBSE
Nonlinear model predictive control
NMPC
Range of motion
ROM
Toronto Rehabilitation Institute
TRI