The Objective Oriented Design of a CUBE Cable - based Parallel Robot
for Arm Rehabilitation Tasks
FOUED INEL1, MOHAMMED KHADEM2, ABDELGHAFOUR SLIMANE TICH TICH2
1Mechanical Engineering Department, Automatic Laboratory,
University of Skikda,
ALGERIA
2Mechanical Engineering Department LGMM Laboratory,
University of Skikda,
ALGERIA
Abstract: - Rehabilitation robots have been employed for training of neural impaired subjects or for assistance
of those with weak limbs. A cube, cable-based parallel robot with eight cables designed for assisting patients in
upper-limb rehabilitation activities, with control over the end-location effector's while locking its rotation
around the horizontal and vertical axes, the device has a lightweight structure that is simple to set up and use
for home usage for both pre-determined and personalized exercises. In this context, we have limited the
tensions of the cables (always positive) and the lengths of the robot do not exceed the workspace. In addition,
the design's kinematic and dynamic studies are presented. The aim of this paper is to help the patient
rehabilitate the upper limb in axes (y-z) and (x-y) with improved patient safety, such that the arm for the patient
can move it in the two planes. The simulation exercises with solidworks and matlab software demonstrate the
effectiveness of our proposed design.
Key-Words: - CBPR- rehabilitation exercises- Kinematic and dynamic model.
Received: May 11, 2022. Revised: August 13, 2023. Accepted: September 16, 2023. Available online: October 31, 2023.
1 Introduction
A cable-based parallel robot (CBPR) is a special
type of parallel robot where cables replace rigid
links. Comparing CBPRs to traditional parallel
robots, this feature gives CDPRs valuable
performances in terms of a large workspace, high
payload, high speed, and acceleration, [1], [2], [3].
The high sensitivity of CBPRs, along with their ease
of installation and reconfiguration, make them
suitable for rehabilitation tasks.in this context,
physical rehabilitation is the process of helping
patients in regaining control over certain limbs
following a protracted sickness or traumatic event,
[4]. Specifically, rehabilitation of the upper limbs.
Additionally, rehabilitation robots are typically
outfitted with sensors that can quantitatively and
continuously monitor the status and progress of each
patient. As a result, various robotic rehabilitation
devices are already accessible, [5], [6]. One of the
most important components of cable-based robots is
the requirement for a suitable control strategy in
order to produce proper movements without
breaking the cables. As mentioned in, [7], [8], the
PD approach was created to increase the robustness
of robotic system control. This PD controller, in
particular, can adjust the control torque depending
on real-time position tracking error in the end-
effector set-point control, [9].
The majority of these robots have a massive
construction and pricey components. To address
these concerns, unique cable-based rehabilitation
activities have been developed. The principal aim of
this work is to help the patient rehabilitate the upper
limb in horizontal and vertical planes with a
maximum distance between the cables and the
patient during the physiotherapy exercises, [10].
The structure of this paper is as follows: firstly
in section one, we introduced a problem
formulation for the design of rehabilitation tasks
with cable based parallel robots. Secondly, we
presented the geometric and dynamic study of our
parallel robot with eight cables. Thirdly, presents
optimization Process and some simulation results
for rehabilitation exercises in the horizontal and
vertical planes. Finally, some conclusions are given
in the last section.
WSEAS TRANSACTIONS on SYSTEMS
DOI: 10.37394/23202.2024.23.1
Foued Inel, Mohammed Khadem,
Abdelghafour Slimane Tich Tich
E-ISSN: 2224-2678
1
Volume 23, 2024
2 The Proposed Design
Figure 1 shows the proposed conceptual design
as based on the fulfillment of specific purposes
or needs. Figure 2a and Figure 2b show
respectively the general geometrical parameters
and the vector analysis that applies to a part of
the robot.
Fig. 1: The proposed design for CUBE - cable based
parallel robot
Fig. 2a: The general geometrical parameters
Fig. 2b: The vector analysis that applies to a part of
the robot
With:
LB: The lengths of the side of the workspace
(LB = 0.65 m).
Li (i=1,..4): the lengths of the cables (Li =0.325
m).
: vector to (ò , o);
: vector to (a , M1);
Li: length of the cable;
R: side length of the robot base.
: vector to (a, ò).
H: height between the base and the motor 8;
Mi: exit point of the cables from the base;
:Unitary matrix ;
: vector to (M1, o);
3 Inverse Geometric Model (IGM)
This section illustrates how to determine the lengths
of the cables "Li", the angles "Qi" between the X,Y
axes and the cables connected to the platform and
"αi" between the Z axis and the planar plane X, Y.
The inverse geometric model can be expressed by
the following equations, [10].
222 )()()( AizzAiyyAixxLi
; (1)
)(arctan Aixx
Aiyy
gi
; (2)
)
)()(
(arctan 22 AiyyAixx
Aizz
gi
; (3)
With: i=1…8.
4 Dynamic Model of the End Effector
In order to analyze the input-output behavior of the
cable-based robot under consideration, we present
the dynamic model, which describes the equation of
motion of the end-effector, [11]:
(4)
Where:
(5)
Is: the vector of the tensions of each cable.
And
x
z
O(0,0,0)
M2
M5
M4
M1
M3
end-effector
cables
M8
M7
M6
Mi
ό(x,y,z)
O(0,0,0)
end-effector
WSEAS TRANSACTIONS on SYSTEMS
DOI: 10.37394/23202.2024.23.1
Foued Inel, Mohammed Khadem,
Abdelghafour Slimane Tich Tich
E-ISSN: 2224-2678
2
Volume 23, 2024
))((),(
X
X
C
Xdt
d
JXSXXN


(6)
The relationship between the applied forces acting
on the end-effector and the cable tensions ti can be
expressed as follows:
tSFR*
(7)
Where:
FR :represents the external forces acting on the end-
effector.
S :is the jacobian matrix.
)()()()(
)()()()()()()()(
)()()()()()()()(
4321
11332211
44332211
ssss
scscscsc
cccccccc
S
)()()()(
)()()()()()()()(
)()()()()()()()(
8765
88776655
88776655
cccc
ssssssss
cscscscs
(8)
with c(θ) and s(θ) represent cos(θ) and sin(θ)
respectively.
We introduce the system states that present the
dynamical model in state space form:
(9)
Where:
rtCJ
(10)
With : ri = r(i=1.2…8).
:is the rotation angle of the pulley (like in
Figure 3).
Fig. 3: The structure diagram of the pulley /shaft
From equation 10:
)(
1
CJ
r
t
(11)
ii LL
LL LL
r
Xi
X
X
0
220
110
2
1
1
)(
)(
)(
(12)
Ci: The viscous damping coefficients of each motor
shaft.
Ji: The inertia of the rotor and the pulley of each
motor.
The state space representation can be derived in the
general form.
)(*),(,)( tUtXgtXFtX
(13)
Where:
)(
0
)(
0
)(
0
)(
3
2
1
tu
tu
tu
tU
(14)
X(t) represents the state space vector, while
F(X, t), g(X, t) are nonlinear functions and U(t)
represents the command vector. The resulting
tension at the end effector leads it to move towards
the required position on its workspace. However, to
work properly, an additional constraint should be
fulfilled concerning the dynamical equilibrium of
the end-effector. This means that, in order to prevent
any cable from collapsing, all the cables should be
maintained under minimal and positive tensions.
5 Optimization Process
Workspaces are one of the kinematic aspects of
parallel robots. All positions accessible to the end-
effector enable the construction of the workspace,
which may be characterized in most cases by an
WSEAS TRANSACTIONS on SYSTEMS
DOI: 10.37394/23202.2024.23.1
Foued Inel, Mohammed Khadem,
Abdelghafour Slimane Tich Tich
E-ISSN: 2224-2678
3
Volume 23, 2024
association of basic geometric models, [12], [13].
The workspace can then be mathematically
expressed and incorporated into the expression of an
objective function. The aim function can be coupled
with and manage an additional kinematic
characteristic known as singularities distribution. In
this formulation, we address the design challenge of
a CUBE cable-based parallel robot for a particular
workspace with a minimum tension that is always
positive throughout all cables.
Figure 4 shows the proposed approach for
rehabilitation exercises in two planes, horizontal and
vertical axes, [14]. Furthermore, it consists of three
different parts: the genetic algorithm represented by
the PID controller, the tension calculation in order
always positive and pulley angle β to determine the
cable lengths Li.
Fig. 4: Optimal design approach for cube- cable
based parallel robot for prescribed workspace
6 Simulation Exercises
Based on the geometric concept with solidworks
software, shown in Figure 5 (a,b and c) , a first
Concept design of the structure has been constructed
to produce a prototype. Furthermore displayed is the
shortest distance between the cables and the patient
during the rehabilitation exercise, [15]. Figure 6
illustrates Flexion of the upper arm in the vertical
plane and Figure 7 shows the movement that did by
the patient. Figure 8 and Figure 9 present
respectively the lengths and the tensions necessary
to do rehabilitation tasks in the vertical plane. The
same way for the horizontal plane.
(a)
(b)
(c)
Fig. 5(a.b.c): Criteria for safety in vertical
rehabilitation tasks.
Boundary intervals
CUBE Cable based parallel robot
The length of cables do not
exceed the workspace
Always-positive cable tension
The vector of the
tensions of each cable
τ = [τ1, τ2, τ3,…, τ8]T
Predefined trajectory
( Motion )
Optimal Design: the command vector
U(t) = [0
)(
1tu
0
)(
2tu
0
)(
3tu
]T
PID
WSEAS TRANSACTIONS on SYSTEMS
DOI: 10.37394/23202.2024.23.1
Foued Inel, Mohammed Khadem,
Abdelghafour Slimane Tich Tich
E-ISSN: 2224-2678
4
Volume 23, 2024
Fig. 6: Flexion of upper arm in vertical plane
Fig. 7: The movement that did by the patient
In the vertical plane.
Fig. 8: The lengths necessary to do rehabilitation
tasks in vertical plane
Fig. 9: The tensions necessary to do rehabilitation
tasks in vertical plane
(a)
(b)
(c)
(d)
Fig. 10 (a.b.c.d): Criteria for safety in horizental
rehabilitation tasks.
90°
z
y
WSEAS TRANSACTIONS on SYSTEMS
DOI: 10.37394/23202.2024.23.1
Foued Inel, Mohammed Khadem,
Abdelghafour Slimane Tich Tich
E-ISSN: 2224-2678
5
Volume 23, 2024
Fig. 11: Flexion of upper arm in horizontal plane
Fig. 12: The movement that did by the patient
In the horizontal plane
Fig. 13: The lengths necessary to do rehabilitation
tasks in horizontal plane
Fig. 14: The tensions necessary to do rehabilitation
tasks in horizontal plane
The criteria for safety in horizental rehabilitation
tasks are presented in Figure 10 whereas, the flexion
of upper arm in horizontal plane is presented in
Figure 11. Similarly, the movement in the horizontal
plane of the patient is presented in Figure 12.
Lastly, it is worth mentioning that Figure 13
presents the necessary lengths to do rehabilitation
tasks in horizontal plane whereas, Figure 14
showcases the tensions necessary to do
rehabilitation tasks in horizontal plane.
The simulation rehabilitation activities in
horizontal and vertical planes illustrate the
best configuration design from the side of the
largest workspace, patient safety and the
performance of cable movements and the
tensions necessary to do rehabilitation
movement into both planes. This last, to
always maintain the position of the cables in
their taut, we are limited to the tensions with
t_max (2N ) and t_min ( 0N ) in order to
always be positive.
7 Conclusion
The topology of a cube, cable-based parallel robot
(CBPR) is the topic of this research. This paper
exposes CUBE, for cable-based parallel robot for
the assistance of patients in rehabilitation
exercising. The main movements involved vertical
and horizontal planes. In this way, we have taken in
account the positive of tensions which are limited
with t_min and t_max and also patient motions have
been measured in order to identify a safe workspace
( the cables do not exceed the work space ) required
for a rehabilitation. Simulation rehabilitation
exercises show the best candidate for rehabilitation
exercise purposes such as large workspace, re-
configurable architecture, and portability
effectiveness.
References:
[1] Marc Gouttefarde & Tobias Bruckmann
‘Cable-Driven Parallel Robots’ Encyclopedia
of Robotics pp 1-14, February 2022.
[2] C. Butefisch, H. Hummelsheim, P. Denzler,
and K. H. Mauritz, “Repetitive
training of isolated movements improves the
outcome of motor rehabil-
itation of the centrally paretic hand,” J.
Neurol. Sci., vol. 130, no. 1, pp. 59–68, 1995.
[3] M. L. Aisen and H. I. Krebs, “The effect of
robot-assisted therapy and rehabilitative
training on motor recovery following stroke,”
WSEAS TRANSACTIONS on SYSTEMS
DOI: 10.37394/23202.2024.23.1
Foued Inel, Mohammed Khadem,
Abdelghafour Slimane Tich Tich
E-ISSN: 2224-2678
6
Volume 23, 2024
Stroke, vol. 28,
pp. 443–446, 1997.
[4] Cafolla, D., Russo, M., & Carbone, G. (2019).
CUBE, a Cable-driven Device for Limb
Rehabilitation. Journal of Bionic Engineering,
16(3), 492–502.
[5] Major K A, Major Z Z, Carbone G, Pîslă A,
Vaida C, Gherman B, Pîslă D L. Ranges of
motion as basis for robot-assisted poststroke
rehabilitation. Human and Veterinary
Medicine, 2016, 8, 192–19.
[6] Sheng B, Meng W, Deng C, Xie S Q. Model
based kinematic & dynamic simulation of 6-
DOF upper-limb rehabilitation robot.
Proceedings of Asia-Pacific Conference on
Intelligent Robot Systems (ACIRS), Tokyo,
Japan, 2016, 21–25.
[7] Tanev T K, Cammarata A, Marano D, Sinatra
R. Elastostatic model of a new hybrid
minimally-invasive-surgery robot. The 14th
IFToMM World Congress, Taipei, Taiwan,
2015.
[8] Inel. F, M. Khadem, A. Slimane Tich Tich.’
Realization and Simulation a Novel Kind of
Parallel Cables-Based’ WSEAS Transactions
on Circuits and Systems, p. 277-284, Vol. 21,
2022,
https://doi.org/10.37394/23201.2022.21.30.
[9] Inel. F, Bouchmal, B. and Khochmane, L.
"Modeling and control of new model in a
spatial coordinates -3D- for cable-based
robots", World Journal of Engineering, Vol.
12 No. 2 (2015), pp. 189-200.
[10] Williams, R.L., P. Gallina, and A. Rossi.
Planar cable-direct-driven robots, part i:
Kinematics and statics. In Proceedings of the
2001 ASME Design Technical Conference,
27th Design Automation Conference. 2001.
[11] Williams II, R.L. Parallel robot projects at
ohio university. In Workshop on fundamental
issues and future research directions for
parallel mechanisms and manipulators,
Quebec City, Canada. 2002.
[12] Loureiro R, Amirabdollahian F, Topping M,
Driessen B, Harwin W. Upper limb robot
mediated stroke therapy GENTLE/s
approach. Autonomous Robots, 2003 15, 35–
51.
[13] Lamine H, Laribi M A, Bennour S, Romdhane
L, Zeghloul S. Design study of a cable-based
gait training machine. Journal of Bionic
Engineering, 2017, 14, 232–244.
[14] Pusey J, Fattah A, Agrawal S, Messina E,
Jacoff A. Design and workspace analysis of a
6–6 cable-suspended parallel robot.
Proceedings of IEEE/RSJ International
Conference on Intelligent Robots and
Systems, Las Vegas, USA, 2003.
[15] Laribi, M. A., Carbone, G., & Zeghloul, S.
(2019). On the Optimal Design of Cable
Driven Parallel Robot with a Prescribed
Workspace for Upper Limb Rehabilitation
Tasks. Journal of Bionic Engineering, 16(3),
503–513.
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
- Foued Inel: Conceptualization, prototyping and
testing
- Mohammed Khadem: methodology, prototyping
and testing, supervision
- Abdelghafour Slimane Tich Tich: testing,
supervision.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
No funding was received for conducting this study.
Conflict of Interest
The authors have no conflict of interest to declare.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on SYSTEMS
DOI: 10.37394/23202.2024.23.1
Foued Inel, Mohammed Khadem,
Abdelghafour Slimane Tich Tich
E-ISSN: 2224-2678
7
Volume 23, 2024