Mechatronics 52 (2018) 70–77

Contents lists available at ScienceDirect

Mechatronics
journal homepage: www.elsevier.com/locate/mechatronics

Model-based tracking control design, implementation of embedded digital T

controller and testing of a biomechatronic device for robotic rehabilitation☆
⁎
Alessio Merola, Domenico Colacino, Carlo Cosentino , Francesco Amato
School of Computer and Biomedical Engineering, Universitá degli Studi Magna Græcia di Catanzaro, Campus Universitario di Germaneto, Catanzaro 88100, Italy

A R T I C LE I N FO A B S T R A C T

Keywords: In this paper, the tracking control problem of a biomimetic exoskeleton powered by a pair of pneumatic artificial
Model-based control muscles is considered. The antagonistic configuration of the pair of pneumatic muscles, which is biologically
Trajectory tracking control inspired, enables safe and reliable actuation in applications of orthopaedic rehabilitation. However, during the
Pneumatic artificial muscles inflation-deflation process, the pneumatic muscles introduce nonlinearity and hysteresis which deteriorate the
Rehabilitation robotics
control performance. A model of the antagonistic artificial muscles is adopted to develop a computed-torque
control for feedforward compensation of the nonlinear dynamics of the actuated joint. A PID control action is
used in combination with the feedforward compensation to achieve fast and accurate tracking control perfor-
mance. The model, which possesses a reduced set of parameters as functions of the inflation/deflation phase,
enables efficient nonlinear compensation. The experimental tests on the biomechatronic device, compared with
other state-of-the-art approaches for controlling pneumatic artificial muscles, show better tracking performance
in terms of convergence rate and robustness, justifying the convenience of using the proposed control metho-
dology in the design of tracking controllers for exoskeletal biomechatronic devices.

1. Introduction Some control techniques have been developed to solve the problems
of regulation and trajectory tracking for PAMs-driven robots. Prior re-
The emerging field of soft robotics is currently covering novel ap- sults focus on variable structure control [8], gain scheduling [9],
plications in rehabilitation robotics, prosthetics and surgical robotics adaptive backstepping [10], sliding mode [11] and PID neural network
and, more in general, several safety-critical applications involving in- control [12]. The sliding mode control strategy recently proposed in
teractions between robots and human operators. The main require- [13] is aimed at the enhancement of the safety during collision with
ments for safe human-robot interactions [1] can be fulfilled through the obstacles. Therefore, the sliding mode tracking controller is com-
inherent (and adaptable) compliance of soft actuators. Moreover, the plemented by a joint compliance controller, which meets the safety
adaptation mechanisms of the compliance of soft robotic and bio- requirements during collision.
mechatronic devices can mimic the behaviour of the biological mus- Model-based compensation strategies are originally proposed in
culo-skeletal system [2,3]. [14,15], where the compensation of the hysteresis in the force char-
As an example of biomimetic and soft actuation, pneumatic artificial acteristics of pneumatic muscles is achieved on the basis of generalized
muscles (PAMs) have been employed in the realization of rehabilitation models of the hysteresis in the mechanical response of PAMs. A feed-
robots, wearable exoskeleton robots and energy-efficient walking hu- forward compensation is implemented into the feedback control
manoids (see [4–6]). More recently, pneumatic muscle actuation tech- schemes of linear positioning stages implementing backstepping [15]
nologies are developed towards the realization of miniaturized bio- and cascade control [14] strategies. More recently, in [16,17], the au-
mechatronic devices. For instance, the work [7] focuses on the thors show how the guaranteed-cost control approach can be effectively
characterization of pneumatic muscles for set-point regulation of the applied to the solution of a regulation problem for a PAMs-driven robot
motion of a biomechatronic finger. whose dynamics can be described through uncertain (bilinear and
Together with their advantages, PAMs offer some challenges in the quadratic) systems.
design and implementation of the tracking control, since the controller To foster the efficient implementation of model-based control stra-
has to handle the strong nonlinearity of the PAMs dynamics. tegies for PAMs-driven robots, lumped parameter models of the

☆
This paper was recommended for publication by Associate Editor Prof Kong Kyoungchul.
⁎
Corresponding author.
E-mail addresses: [email protected] (A. Merola), [email protected] (D. Colacino), [email protected] (C. Cosentino), [email protected] (F. Amato).

https://doi.org/10.1016/j.mechatronics.2018.04.006
Received 12 June 2017; Received in revised form 16 February 2018; Accepted 25 April 2018
0957-4158/ © 2018 Published by Elsevier Ltd.
A. Merola et al. Mechatronics 52 (2018) 70–77

nonlinear dynamics of PAMs can be exploited. The three-element

phenomenological model proposed by Reynolds et al. [18] allows to
accurately predict the dynamic response of PAMs; the model para-
meters can be easily identified through static perturbation tests at
several constant values of pressure. More recently, in [19] a series of
static perturbation tests have been carried out for the efficient identi-
fication of a novel model of the mechanical response of some classes of
PAMs.
The main contribution of this work consists in the development and
the experimental evaluation of a novel model-based tracking control
methodology for a biomechatronic device powered by PAMs in antag-
onistic configuration. The lumped parameter model by Reynolds et al.
[18] is used here to develop a computed torque control for feedforward
compensation of the actuated joint via nonlinear inversion. The Rey-
nolds’s model provides an efficient description of the variability of the
nonlinear dynamics of the PAMs during the pressure cycle. Therefore,
the feedforward control action is a function of the model parameters
which depend on the phase of inflation/deflation. The compensation
loop of the tracking controller is complemented by a PID control action
which enables the robust regulation to zero of the tracking error.
To the best of the authors’ knowledge, this is the first time that such
a combined nonlinear inversion feedforward + feedback PID control
methodology has been proposed and experimentally tested for the
tracking control of a PAM-based robot. The devised approach improves
the tracking performance over the existing approaches in the related Fig. 1. Overview of CoRAnT. (a) Orthotic shoe. (b) Pneumatic muscle actuator.
(c) Pressure regulator. (d) Load cell. (e) Optical encoder.
literature on the control of PAMs-based robots. The parsimony of the
model makes this control approach very suitable for the real-time im-
plementation on embedded microcontroller devices. Therefore, a non To obtain a training performance at least comparable to the manual
negligible byproduct of this work is the microcontroller-based im- therapy, an essential requirement for biomechatronic rehabilitation is
plementation of the embedded control system, whereas the previously the safe interaction between the patient and the rehabilitation device.
published results were tested by means of PC-based implementations. During the conventional clinical therapy, the manipulations performed
Some experimental tests are presented to prove the superior control by the therapist are adapted to the resistance force exerted by the pa-
performance achieved by the proposed methodology thanks to the ef- tient. Similar adaptation mechanisms, guaranteeing the intrinsic safety
ficient compensation of the nonlinear dynamics. Specifically, the per- of robotic rehabilitation tasks, can be realized through suitable control
formance of the closed loop control system has been measured in terms strategies of biomechatronic devices driven by biomimetic and soft
of convergence rate, steady-state error and robustness to load dis- actuators. For instance, the natural compliance of soft actuators, which
turbance during the tracking of constant and sinusoidal trajectories. utilize air as source of energy, enables to absorb potential shocks oc-
Furthermore, the micro-controller based implementation of the pro- curring during the manipulation of the patient.
posed control scheme shows the advantages related to the real-time After the analysis of both the main issues involved in physical re-
execution, computational resources and customizable constraints on the habilitation and the advantages provided by robotic and biomecha-
control action. tronic technologies, the main specifications are implemented in a ro-
This paper is structured as follows. The main technical specifica- botic exoskeleton for rehabilitation following the design principles of
tions of the robotic exoskeleton and a model of the dynamics of the soft and biomimetic robotics. Therefore, CoRAnT is actuated by soft
actuated robot are presented in Sections 2 and 3, respectively. The pneumatic muscles, which guarantee safety and comfort for the patient.
model-based tracking control law is derived in Section 4, where the The mechanical structure and the main components of the robotic
steps of design and implementation of the digital controller on em- exoskeleton are highlighted in Fig. 1; the pneumatic muscle used for the
bedded control unit are also described. The experimental tests and the actuation of the exoskeleton is the fluidic muscle MAS-20-200N by
discussion of the experimental results are given in Section 5. Some FESTO company.
concluding remarks are eventually left to Section 6. An orthotic shoe is installed on the rotational joint of the exoske-
leton; the joint axis is aligned with the ankle axis of the patient (see
2. CoRAnT: Compliant Robotic Ankle Trainer Fig. 2).
A couple of PAMs is required for the full actuation of the joint, since
The traditional therapy for stroke and traumatic brain injuries re- a single PAM actively generates motion in one direction. The PAMs are
quires long and intensive rehabilitation tasks performed by the physical arranged in the bio-inspired configuration of Fig. 3, where the antag-
therapist. Indeed, the quick recovery of neural and motor functions, onistic setup can mimic the mechanisms of regulation of motion and
resulting from cortical reorganization in the motor cortex of the stiffness of the joints in the human musculo-skeletal system. Therefore,
patient, can be achieved under intensive and repetitive exercises. in analogy with the human anatomy, the PAMs act as biceps and tri-
Unfortunately, the required burden of care may contrast with the ceps, respectively. The conversion of the linear motion of the actuators
budget and time constraints. to the joint rotation is obtained through some cables of diameter
Thanks to the accuracy and repeatability of the control perfor- 3.5 mm tied together around a pulley of diameter 80 mm.
mance, robotic exoskeletons performing high-intensity physical re- Position and force sensors are installed on the robotic exoskeleton.
habilitation can offer their potential in the optimization of costs and An optical rotary encoder (AVAGO HEDM5500 J14) measures the an-
efficiency of the rehabilitation procedures. Moreover, biomechatronic gular position of the joint at 1024 counts per revolution. The encoder
and robotic devices can collect quantitative data useful for the eva- provides the measured variable to the tracking controller. Moreover,
luation of both the patient’s motor performance and the progress of the each PAM is connected to a load cell (Phidgets 3138 S-type). The signals
motor recovery. of force and angle can be used for measuring the interaction between

71
A. Merola et al. Mechatronics 52 (2018) 70–77

Table 1
Model parameters.
Parameter Value

Spring k0 4.82 [N/m]

element k1 131 [N/m/bar]
Damping b0i (b0d) 1.22 (1.31) [N/m/s]
element b1i (b1d) 135 (−230) [N/m/s/bar]
Force f0 14.89 [N]
element f1 2090 [N/bar]

sense of the contraction displacement.

The nonlinearity introduced by the PAMs is described through the
three-element phenomenological model by Reynolds et al. [18]. The
expression of the total force ϕ exerted by a single PAM is obtained from
a parallel configuration of an elastic element (spring) of stiffness k(p), a
viscous element (damper) of damping coefficient b(p) and a contractile
element generating the active force contribution f(p). Therefore, the
PAM force reads
ϕ = f (p) − b (p) s˙ − k (p) s,

where p denotes the pressure at which the PAM is inflated and s is the
contraction length. The model parameters follow the constitutive laws
Fig. 2. CoRAnT performing a rehabilitation task on the patient. (a) Lateral f (p)= f0 + f1 p [N]
view. (b) Front view.
b0i + b1i p [N/m/s] inflation
b (p)= b0 + b1 p = ⎧
⎩ b0d + b1d p [N/m/s] deflation
⎨

k (p)= k 0 + k1 p [N/m]

The values of the parameters of the damping function b(p) vary de-
pending on the phase of inflation/deflation.
Concerning the values of the model parameters, we refer both to
Table 1 and to the identification procedure described in [18]. The va-
lues listed in Table 1 are taken from [13], where the model parameters
have been identified for the same FESTO fluidic muscles adopted in
CoRAnT. The parameters of the elastic and damping functions of the
pressure can be identified from the step response of the actuator at
constant pressure and subject to a step change of the applied load. The
contractile (active force) element function is determined by a step
change to the inflation pressure of the actuator under constant load.
The model of the actuated joint requires the moment of inertia of
CoRAnT. The estimated moment of inertia amounts to Il = 0.0023
kg m2. This value can be obtained from the CAD 3D model of CoRAnT
after specifying the materials of the parts of the model.
Fig. 3. Actuation scheme of CoRAnT. The torque of the actuated joint reads
τ = τb − τt = (ϕb − ϕt ) r , (1)
the robot and the patient, e.g., through the measurement of the torque
where τb and τt are the torques generated by each PAM, as in
opposed by the patient during the motion cycle. For instance, the pa-
tient’s participation to the rehabilitation task could be estimated τb= (fb − bb s˙b − kb sb) r (2a)
through the measured data.
The regulation of the pressure of the PAMs, involved in the trajec- τt = (ft − bt s˙t − kt st ) r . (2b)
tory control of the joint, is obtained through two proportional pressure
In (2), sb and st denote the contraction length of biceps and triceps,
regulators FESTO MPPE-3-1/8-6-010-B. The implementation of the
respectively. From the antagonistic configuration of Fig. 3, a positive
pressure regulation into the motion control system of the exoskeleton is
rotation is obtained through the active force generated by the inflation
discussed in detail in Section 4.
of the biceps, whereas the triceps deflates without active contribution
to motion. For a negative rotation, it is required that the triceps gen-
3. Derivation of the equations of motion including actuator erates the force driving the joint rotation, whereas the biceps is passive.
dynamics The total torque can be expressed in terms of the joint angle θ, using
the kinematic relations
Starting from the kinematic scheme of Fig. 3, the equations of mo-
π π
tion of CoRAnT are derived. sb = r ⎛θ + ⎞, st = r ⎛ − θ ⎞.
⎝ 6⎠ ⎝6 ⎠
Following the notation of Fig. 3, clockwise rotations are positive,
whereas the upright vectors ϕb and ϕt denote the forces exerted by bi- In a first stage of the actuation strategy, each PAM is inflated at the
ceps and triceps, respectively. In Fig. 3, the force vectors point upward initial positive pressure p0. Thereafter, the joint rotation is obtained
since each PAM generates its (actuation or resistance) force only in the through the antagonistic control of the pressure of the pair of PAMs. A

72
A. Merola et al. Mechatronics 52 (2018) 70–77

positive/negative rotation of the joint requires an increment/decrement approach of the computed-torque control (see [20], Ch. 6.6). The
of the pressure pb of the biceps and a decrement/increment of the control input can be computed in real time according to the feedfor-
pressure pt of the triceps. Therefore, the antagonistic actuation can be ward-feedback scheme, after measuring both the kinematic variables
formulated as and the model parameters of the actuated joint.
pb = p0 + Δp , pt = p0 − Δp , The tracking control system, following a time-varying reference
trajectory θd(t), requires the derivation of the tracking error dynamics
where p0 is the initial pressure and Δp is the pressure difference from p0 as a function of the error e : =θd − θ and its successive derivative
used as manipulated variable in the antagonistic pressure control. It is e˙: =θ˙d − θ˙ and e¨: =θ¨d − θ¨ .
useful to note that the antagonistic strategy simplifies the motion The nonlinear compensation is achieved through the control input
control, since only one control variable, the pressure difference Δp, is
Il θ¨d + αθ + βθ˙ + τg (θ) + v
used for the actuation of the joint through two PAMs. u= ,
Therefore, the dynamics of the biomechatronic joint is described as δ + σθ˙ (6)
where v is an auxiliary control signal to be designed in order to achieve
Il θ¨ + ζθ˙ + τg (θ) + d = τ , (3)
robust tracking performance. To this end, v is taken as a PID control
where Il denotes the link inertia, ζ is the viscous friction coefficient at action in the form
the joint, τg(θ) denotes the gravity torque, τ is the control input torque
and d represents the unknown external input disturbance which embeds
v = kP e + kV e˙ + kI ∫ edt.
some external perturbations, e.g., load variations or disturbance torque, Let us define the state vector as
other than modelling mismatches, unknown parametric uncertainties,
friction nonlinearities, etc. ⎡ ∫ edt ⎤
Since the PAMs actuating the joint are identical, the spring element x: =⎢ e ⎥,
⎣ e˙ ⎥
⎢ ⎦
parameters are taken as k 0b = k 0t = k 0 and k1b = k1t = k1. The pressure
difference Δp is the control input of the closed-loop system, i.e. u = Δp. then the state space representation of the closed loop system yields
After introducing in (3) the constitutive laws of the actuators, the
equation of motion of the joint, including actuator dynamics, reads x˙ = Ax + Bw,
where
Il θ¨ + (ζ + β ) θ˙ + αθ + τg (θ ) + d = (δ + σθ˙ ) u, (4)
with ⎛ 0 1 0 ⎞ ⎛0 ⎞
A=⎜ 0 0 1 , B = ⎜ 0 ⎟,
⎜−I −1 k −I −1 k −I −1 k ⎟⎟ ⎜ I −1⎟
α = 2r 2 (k 0 + k1 p0 ) (5a) ⎝ l I l P l V ⎠ ⎝l ⎠

β = r 2 [b0b + b0t + p0 (b1b + b1t )] (5b) and w defines the disturbance which embodies friction effects, model
uncertainties and unknown interaction torque.
π The nominal value of the inertia and the measured values of θ and θ̇
δ = 2rf1 − k1 r 2
3 (5c) are introduced in the control action (6). Moreover, the value of the
σ = r 2 (b1t − b1b). parameters α, β and δ are computed offline after the regulation of the
(5d)
initial pressure to the desired value p0; only the parameter σ, which
Given the initial pressure p0, all the parameters defined in (5) do not depends on the inflation/deflation phase, requires an online computa-
vary during the motion, except for the value of σ which depends on the tion.
phase of inflation/deflation. The PID controller parameters, as shown in Table 2, are tuned and
The parameter α can be associated to a stiffness coefficient of the verified experimentally for optimal results.
elastic term αθ arising from the spring element of the three-element The tracking controller is implemented on 8-bit AVR RISC-based
phenomenological model of the PAM, whereas β can be viewed as an microcontroller ATmega 328; the control signal and the measurements
equivalent viscous friction factor depending on both the initial pressure are processed at 50 Hz. An I2C bus interconnects all the digital com-
and damping effects of the actuator. The parameters δ and σ describe ponents of the embedded control system: microcontroller, optical en-
the variability of the generated torque through the damping and active coder and digital-analog converter (DAC).
force elements of the actuator model. The pressures of the PAMs are regulated through two proportional
pressure regulators in the range 0–6 bar. Two complementary control
4. Model-based tracking control and digital implementation signals are generated to drive the pressure regulators associated to the
pair of antagonistic muscles. One signal inflates/deflates the biceps and
A satisfactory trajectory-tracking control performance can be the other one deflates/inflates the triceps in order to produce positive/
achieved through the combined feedforward-feedback control scheme negative rotations of the joint. The control signals in the range 0–10 V,
of Fig. 4, which takes into account the nonlinearity introduced by the which correspond to the pressure outputs 0–6 bar, are obtained after
pneumatic actuators. A control law, which attains a compensation of conversion of the control signals via a dual channel 12-bit DAC
the nonlinear terms of the dynamics (4), is developed through the TLV5618 and through amplification by a programmable

Fig. 4. Torque inverse control scheme.

73
A. Merola et al. Mechatronics 52 (2018) 70–77

Table 2 feedforward-feedback PID control scheme: the control action is char-

Parameters of the PID controller. acterized both by smoothness in response to sudden changes in desired
Parameter Value position and by robustness against the patient load.
The experimental results show that the proposed inverse model-
KP 30.1 [bar/deg] based controller is capable of efficiently controlling the antagonistic
KD 2.05 [bar s/deg]
mechanism of the PAMs pair, while compensating for nonlinear phe-
KI 1.2 [bar/deg/s]
nomena and external load disturbances occurring during the inflation-
deflation cycle.
instrumentation amplifier INA122. The superior performance achieved by the proposed tracking con-
Based on desirable safety attributes on the interaction between the troller can be appreciated with respect to the related approaches in the
patient and CoRAnT, some constraints on the controller output can be scientific literature on nonlinear control of pneumatic muscle actuators
posed by setting into the control routine a maximum pressure limit. in antagonistic configuration. Our model-based tracking controller
During the trajectory tracking with minimum joint compliance, the outperforms some sliding mode control strategies [11,21] and non-
error from the reference trajectory should be minimized by setting a linear PID control techniques [12,22].
maximum pressure limit compatible with the actuator saturation, In [11], a sliding mode approach is applied to the trajectory
whereas a strategy of maximum compliance suggests to lower such a tracking control of a pneumatic artificial muscle manipulator; a max-
limit in order to safely absorb the involuntary contractions of the pa- imum joint error of 7° is achieved for a sinusoidal reference trajectory.
tient. Chattering, lack of robustness and slow responses are the major issues
In the following tests for the experimental validation of the tracking involved in sliding mode control strategies. In [21], a sliding mode
control performance, the maximum pressure limit is set to 5.7 bar in tracking controller of a pneumatic artificial muscle manipulator is op-
order to improve the tracking accuracy. timized in order to attenuate the oscillations in the system response.
The mitigation of the oscillations is obtained at the cost of a longer
settling time since, for a step change of about 35° in the desired angle,
5. Results and discussion the joint angle converges to the desired value after a minimum transient
of 2.5 s.
The results of the tests, performed on CoRAnT during some re- In [12], a nonlinear PID controller using a neural network is pre-
habilitation sessions, are presented and discussed in this Section. sented for controlling a planar robotic arm actuated by PAMs in an-
During the phase of design of experiments, the range of motion and tagonistic setup. After training of the neural network, a maximum
frequency of the sinusoidal references are selected on the basis of the tracking error of about 2.5° is obtained following a sinusoidal reference
biomechanical requirements for lower limb rehabilitation tasks. The trajectory ranging from −20° to 20° at frequency 0.2 Hz.
trajectory tracking control performance of CoRAnT has been evaluated Further comparison with the nonlinear PID controller of an exos-
using the reference profiles at different frequencies. keletal wrist in [22] shows the superior performance of our feedfor-
A set of tests have been carried out on the robotic exoskeleton ward-feedback PID control scheme in terms of fast and accurate
performing a rehabilitation session on a female subject (29 y.o., weight tracking. To this end, it is useful to compare the step responses of the
53 kg) sitting on a chair. The initial pressure of the PAMs is set to 3 bar; PAM-based exoskeletal wrist, which are presented in [22] for constant
this value provides a limited stiffness of the joint in favour of the safety angular references of −10°, −20°, −30°, 10°, 20° and 30°. The ex-
constraints. periments of flexion-extension movements are performed on the exos-
Fig. 5 shows the results of the sinusoidal trajectory tracking at fre- keleton equipped with a spherical joint simulating a passive wrist; the
quency of 0.5 Hz and 1 Hz, achieving a root mean square error of 1.25° rising time in response to the references of −20° and 20° is of the order
and 1.76°, respectively. of 0.5 s. The step responses acquired on CoRAnT exhibit better tracking
The feedforward action compensating for the nonlinearity of the performance when compared to the respective responses of the exos-
actuators is essential to guarantee good tracking performance, as shown keletal wrist, since the absolute tracking error reduces to 0.4° after
from the results of Fig. 6 where the tracking error is compared for two 0.28 s in presence of the patient load.
different control configurations. The tracking control performance achieved during the flexion-ex-
For the same sinusoidal reference with amplitude of 15° and fre- tension experiments in [22] has been evaluated for sinusoidal reference
quency of 0.5 Hz, the first configuration implements a PID without trajectories at maximum frequency of 0.4 Hz. The execution frequency
model-based compensation; in the second one, the tracking perfor- of the PID controller is 10 Hz, a value compatible with the time re-
mance is improved considerably by adding to the pure feedback scheme quired for computing the control action.
the nonlinear compensation law of Section 4. The improvement of the control performance by the feedforward-
The aforementioned experimental results show that CoRAnT can feedback scheme of CoRAnt is confirmed also for the case of tracking of
perform repetitive rehabilitation tasks through satisfactory tracking a sinusoidal reference over an extended range of frequency, up to 1 Hz.
accuracy. Even in the case of increase in the frequency of the reference Moreover, the model-based control law can be efficiently implemented
trajectory, the good performance of CoRAnT is guaranteed by the at the higher frequency of 50 Hz on embedded microcontroller, without
proposed feedforward-feedback control scheme. In this respect, the fast requiring the PC-based implementations involved in [11,12,21,22].
and accurate tracking can be evaluated from the closed loop response to The experimental results prove that the proposed control scheme
step input. succeeds in guaranteeing the robust and efficient control performance
The tracking error obtained for the experimental step responses is required for rehabilitation purposes and, also in comparison with the
represented in Fig. 7; the constant reference signals used in the setpoint existing results in the related field of the control of PAMs-based exos-
tracking control cover the range of angular displacement allowed by keletons, it yields fast and accurate tracking performance.
mechanical hardware of the robot. The experimental tests, performed
for the cases of flexion (Θref = +15∘) and extension (Θref = −15∘), are 6. Conclusions
characterized by fast convergence to the setpoint, with smooth tran-
sient and limited control effort. In particular, after 0.28 s the absolute This work sets out with the objective of designing efficient strategies
value of the error does not exceed 0.4°. The error at 2 s is negligible and of precision tracking control for a biomechatronic device, using a
it amounts to 0.057° (extension) and to 0.114° (flexion). combined feedforward-feedback control scheme for rehabilitation
Fig. 8 illustrates other important features of the combined tasks.

74
A. Merola et al. Mechatronics 52 (2018) 70–77

Fig. 5. Tracking of sinusoidal trajectory with amplitude range of 20° (reference signal in blue line and measured angle in red dashed line). (a) Frequency of 0.5 Hz.
(b) Frequency of 1 Hz. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6. Tracking error with (blue line) and without (red dashed line) compensation action. (For interpretation of the references to colour in this figure legend, the
reader is referred to the web version of this article.)

Both a model of the PAMs-actuated joint and a nonlinear control The comparison of the experimental results to the respective ones
strategy are provided that enable reasonably accurate compensation of obtained via other nonlinear control techniques for PAMs-based exos-
the nonlinear dynamics while keeping the computational burden low. keletons reveals that the improved control performance can be
The model-based control law, achieving a balance between control achieved in all the scenarios of robotic rehabilitation where fast and
precision and computational requirements, allows the efficient im- robust trajectory tracking is required.
plementation of the tracking control on embedded digital controller. Future works will be directed to the control of the impedance of the
The experimental tests on the rehabilitation exoskeleton, following the robot joint using the viscoelastic properties of the PAMs. Attention will
digital implementation and tuning of the controller on embedded also be devoted to reference trajectory generation, e.g., using profiles
control unit, have shown good tracking performance and robustness. taken from gait analysis data.

75
A. Merola et al. Mechatronics 52 (2018) 70–77

Fig. 7. Setpoint tracking error of flexion (θref = +15∘ ) and extension movements (θref = −15∘ ).

Fig. 8. Commanded pressure on the PAMs pair for (a) flexion (θref = +15∘ ) and (b) extension movements (θref = −15∘ ).

References [6] Vanderborght B, Verrelst B, Van Ham R, Van Damme M, Lefeber D, Duran BMY,
et al. Exploiting natural dynamics to reduce energy consumption by controlling the
compliance of soft actuators. Int J Rob Res 2006;25(4):343–58. http://dx.doi.org/
[1] Ham R, Sugar T, Vanderborght B, Hollander K, Lefeber D. Compliant actuator de- 10.1177/0278364906064566.
signs. Robot Autom Mag IEEE 2009;16(3):81–94. http://dx.doi.org/10.1109/MRA. [7] Oliver-Salazar M, Szwedowicz-Wasika D, Blanco-Ortega A, Aguilar-Acevedo F, Ruiz-
2009.933629. Gonzalez R. Characterization of pneumatic muscles and their use for the position
[2] Repperger DW, Phillips CA, Neidhard-Doll A, Reynolds DB. Power/energy metrics control of a mechatronic finger. Mechatronics 2017;42:25–40.
for controller evaluation of actuators similar to biological systems. Mechatronics [8] Repperger D, Johnson K, Philips C. A VSC position tracking system involving a large
2005;15:459–69. http://dx.doi.org/10.1016/j.mechatronics.2004.10.002. scale pneumatic muscle actuator. Proceedings of of 37th IEEE inter-
[3] Caldwell DG, Tsagarakis N, Medrano-Cerda GA. Biomimetic actuators: polymeric nationalconference on decision and control, Tampa, FL, USA. 1998.
pseudo muscular actuators and pneumatic muscle actuators for biological emula- [9] Repperger D, Johnson K, Philips C. Nonlinear feedback controller design of a
tion. Mechatronics 2000;10:499–530. http://dx.doi.org/10.1016/S0957-4158(99) pneumatic muscle actuator system. Proceedings of 1999 American control con-
00071-9. ference, San Diego, CA, USA. 1999.
[4] Noritsugu T, Tanaka T. Application of rubber artificial muscle manipulator as a [10] Carbonell P, Jiang Z, Repperger D. Nonlinear control of a pneumatic muscle ac-
rehabilitation robot. IEEE/ASME Trans Mechatron 2002;2(4):259–67. tuator: backstepping vs. sliding-mode. Proceedings of of the 2001 IEEE inter-
[5] Xing K., Xu Q., He J., Wang Y., Liu Z., Huang X.. A wearable device for repetitive nationalconference on control applications, Mexico City, Mexico. 2001.
hand therapy. In: Proceedings of 2nd IEEE RAS-EMBS international conference on [11] Lilly J, Quesada P. A two-input sliding-mode controller for a planar arm actuated by
biomedical robotics and biomechatronics, Scottsdale, AZ, USA; 2008. four pneumatic muscle groups. IEEE Trans Neural Syst Rehabil Eng

76
A. Merola et al. Mechatronics 52 (2018) 70–77

2004;12(3):349–59. Domenico Colacino was born in Catanzaro, Italy, in 1984.

[12] Thanh TDC, Ahn KK. Nonlinear PID control to improve the control performance of He received the Laurea degree in Biomedical Engineering
2 axes pneumatic artificial muscle manipulator using neural network. Mechatronics from the University of Catanzaro Magna Græcia, Italy in
2006;16(9):577–87. 2008. He received the Ph.D. degree in Computer and
[13] Choi T-Y, Choi B-S, Seo K-H. Position and compliance control of a pneumatic muscle Biomedical Engineering from the University of Catanzaro
actuated manipulator for enhanced safety. Control Syst Technol IEEE Trans Magna Græcia, Italy, in 2014. He is particularly interested
2011;19(4):832–42. http://dx.doi.org/10.1109/TCST.2010.2052362. in modelling and design of mechatronic devices for bio-
[14] Vo Minh T, Tjahjowidodob T, Ramonc H, Van Brussel H. Cascade position control of medical applications and also in analysis and control of
a single pneumatic artificial muscle-mass system with hysteresis compensation. nonlinear systems.
Mechatronics 2010;20(3):402–14.
[15] Schindele D, Aschemann H. Model-based compensation of hysteresis in the force
characteristic of pneumatic muscles. 2012 12th IEEE international workshop on
advanced motion control (AMC). 2012. p. 1–6.
[16] Amato F, Colacino D, Cosentino C, Merola A. Robust and optimal tracking control
for manipulator arm driven by pneumatic muscle actuators. 2013 IEEE interna-
Carlo Cosentino received the Laurea degree (M.Sc.) in
tional conference on mechatronics (ICM). 2013. p. 827–34.
Computer Engineering and the Ph.D. degree in Computer
[17] Amato F, Colacino D, Cosentino C, Merola A. Guaranteed cost control for uncertain
and Automation Engineering, both from Federico II
nonlinear quadratic systems. European control conference (ECC). 2014. p. 1229–35.
University of Naples, Italy, in 2001 and 2005, respectively.
http://dx.doi.org/10.1109/ECC.2014.6862287.
Since 2014 he is Associate Professor of Systems and Control
[18] Reynolds DB, Repperger DW, Phillips CA, Bandry G. Modeling the dynamic char-
Theory at Magna Græcia University of Catanzaro. He has
acteristics of pneumatic muscle. Ann Biomed Eng 2003;8(3):310–7.
published more than 100 articles in international peer-re-
[19] Takosoglu JE, Laski PA, Blasiak S, Bracha G, Pietrala D. Determining the static
viewed journals, conference proceedings and edited books
characteristics of pneumatic muscles. Meas Control 2016;49(2):62–71. http://dx.
and is co-author of two scientific monographs. His research
doi.org/10.1177/0020294016629176.
interests include finite-time stability of linear systems, sta-
[20] Siciliano B, Khatib O, editors. Springer handbook of robotics. Springer; 2008.
bility of nonlinear quadratic systems and applications of
[21] Van Damme M, Vanderborght B, Verrelst B, Van Ham R, Daerden F, Lefeber D.
systems and control theory to the fields of systems and
Proxy-based sliding mode control of a planar pneumatic manipulator. Int J Rob Res
synthetic biology.
2009;28(2):266–84. http://dx.doi.org/10.1177/0278364908095842.
[22] Andrikopoulos G, Nikolakopoulos G, Manesis S. Design and development of an
exoskeletal wrist prototype via pneumatic artificial muscles. Meccanica
2015;50(11):2709–30. http://dx.doi.org/10.1007/s11012-015-0199-8. Francesco Amato was born in Naples on February 2, 1965.
He received the Laurea and the Ph.D. Degree both in
Electronic Engineering from the University of Naples in
Alessio Merola was born in Catanzaro, Italy, in 1979. He 1990 and 1994 respectively. From 2001 to 2003 he has
received the Laurea degree (summa cum laude) in been Full Professor of Automatic Control at the University
Mechanical Engineering from the University of Calabria, of Reggio Calabria. In 2003 he moved to the University of
Italy, and the Ph.D. degree in Computer and Biomedical Catanzaro, where, since 2010, he is Professor of
Engineering from the University of Catanzaro Magna Bioengineering. He is currently the Dean of the School of
Græcia, Italy, in 2003 and 2008, respectively. He is cur- Computer and Biomedical Engineering, the Coordinator of
rently Assistant Professor of Systems and Control the Doctorate School in Biomedical and Computer
Engineering at the Department of Experimental and Clinical Engineering, the Director of the Biomechatronics
Medicine of the University of Catanzaro Magna Græcia. He Laboratory, and member and vice-president of the
is coauthor of about 40 scientific papers, published on in- Concilium of the School of Medicine and Surgery. The sci-
ternational journals and proceedings of international con- entific activity of Francesco Amato has developed in the fields of systems and control
ferences. His current research interests include analysis and theory with applications to the contexts of the computational biology and of the modeling
control of nonlinear systems and control of biomechatronic and control of biomedical systems. He has published about 250 papers in international
devices. Journals and conference proceedings and two monographies with Springer Verlag en-
titled “Robust Control of Linear Systems subject to Uncertain Time-Varying Parameters”
and “Finite-Time Stability and Control”.

77