Lower limp exoskeleton

LOWER LIMB EXOSKELETON
Made by:
Mohamed Said Helmy El Hefnawy
Mohamed Ashraf Mohamed Daoud
Osama Maher Abdul Alim Ibrahim
Amr Mousa Hassan Abdel Wanis
Under the supervision of:
Prof. Dr. Azza Fathallah Barakat
Dr. Aya Abd Allah
This project report is submitted in partial fulfilment of the
requirements for the award of the degree of
Bachelor of Engineering
Department of Mechanical Engineering
Mechatronics Division
Faculty of Engineering, Helwan university
July, 2022

ii
DECLARATION
I hereby declare that this project report is based on our original work except for
citations and quotations which have been duly acknowledged. I also declare that it has
not been previously and concurrently submitted for any other degree or award at FEHU
or other institutions.
Signature : _________________________
Name :
ID No. :
41223037
41223034
41223008
41223027
Date : 12/07/2022

iii
APPROVAL FOR SUBMISSION
I certify that this project report entitled “LOWER LIMB EXOSKELETON” was
prepared by
has met the required standard for submission in partial fulfilment of the requirements
for the award of Bachelor of Mechanical Engineering at Helwan University.
Approved by,
Signature : _________________________
Supervisors :
Prof. Dr. Azza Fathallah Barakat
Dr. Aya Abd Allah
Date : 12/07/2022

iv
ACKNOWLEDGEMENTS
I would like to thank everyone who had contributed to the successful
completion of this project. I would like to express my gratitude to our project
supervisors Prof. Dr. Azza Fathallah Barakat & Dr. Aya Abd Allah for their
invaluable advice, guidance, and their enormous patience throughout the
development of the project.
In addition, I would also like to express my gratitude to Dr. Mohamed Abd
el Ghani for assisting us and facilitating the usage of spare components for the
project.
Also, Eng. Ibrahim badawy for his constant support throughout the year and
helpful insights and recommendations.
And the Academy of scientific research for sponsoring this project and
suppling the necessary funds to see it through.
Finally, the Technicians in our university workshops for the assistance in the
project’s manufacturing and assembly .

v
LOWER LIMB EXOSKELETON
ABSTRACT
The lower limb exoskeleton is an external non surgically invasive assistive device. In
our use case it’s aimed at paraplegics/spinal cord injury (SCI) patients, replacing the
function of the paralyzed/atrophied limbs. Our design is a 6 DOF design powered via
4 DC brushed motors at the hip & knee joints, with two spring loaded ankle joints.
The two mechanical limbs are attached to the patient via straps supporting his/her
weight and providing actuation. The motion trajectories of the hip and knee joints is
fully automated and provided by the control algorithm to approximate the dynamics
overtime While currently it still requires the use of 2 crutches for stability, just so the
design can as slim as possible. Providing fully natural walk gait cycle, standing up &
sitting down would be easily attained as well additionally stair climbing &
descending mode is currently in the future plans for this project.
Mode switching is either done would be done with image recognition provided via a
small camera strapped to the chest OR manually via a toggle switch on one of the
crutches. With the Intention of motion provided also by a button on one of the
crutches, as it initiates the gait cycle from the starting position with a future plan to
implement automatic initiation of the gait cycle via detecting intention of the patient,
which would be done via the use of ultrasonic sensors detecting a specified distance
between the crutch and the limb, but currently the intention of motion is sent
manually by the used via push button.
Keywords: exoskeleton robot; muscle driven simulator; spinal cord injury; PID;
augmentation; assistive devices; robotics; wearable robot; human–robot interaction

vi
TABLE OF CONTENTS
DECLARATION ii
APPROVAL FOR SUBMISSION iii
ACKNOWLEDGEMENTS iv
ABSTRACT v
TABLE OF CONTENTS vi
LIST OF TABLES ix
LIST OF FIGURES ix
CHAPTER
INTRODUCTION 1
1.1 Background 1
1.2 Problem Statements: 1
1.3 Aims and Objectives: 2
LITERATURE REVIEW 4
2.1 Current & past products/research prototypes 4
2.2 Gait cycle & Modelling 6
2.2.1 Weight acceptance (0-12%): 6
2.2.2 Single limb support (12-50%): 7
2.2.3 Swing phase (50-100%): 7
2.3 Muscular Forces 9
2.4 Dynamic model 11
2.5 Control strategies 12
METHODOLOGY 14
3.1 Trajectory planning 14

vii
3.2 Inverse Kinematics & Dynamics 17
3.3 Design 26
3.3.1 The Design main idea & procedure: 26
3.3.2 The Design Configuration: 28
3.3.3 Knee Joint: 30
3.3.4 Ankle Joint: 30
3.3.5 Man–Exoskeleton coupling: 32
3.3.6 Torso coupling: 33
3.3.7 Foot coupling: 33
3.3.8 Links coupling: 35
3.4 Material selection 37
3.4.1 Links Material: 37
3.4.2 Joint Material: 39
3.5 Control strategy 40
RESULTS AND DISCUSSIONS 42
4.1 Simulation 42
4.1.1 Control model /Simscape multibody 43
4.2 Simulink block diagram 46
4.2.1 Hold and shift signal: 47
4.2.2 Signal processing: 48
4.2.3 PID control 49
4.2.4 PWM block 50
4.2.5 GPIO write 51
4.2.6 GPIO read 52
4.2.7 Encoder block 52
4.3 Components Used 53
4.3.1 Electrical Components 53
4.3.2 Mechanical Components 57
4.3.3 Electronic Components 60
4.4 Circuit connections 64
4.5 Assembly 65

viii
CONCLUSION AND RECOMMENDATIONS 68
4.6 Limitations 68
4.7 Future work/recommendations 68
4.8 Conclusion 69
4.9 Closing words 69
REFERENCES 70
APPENDICES 72

ix
LIST OF TABLES
TABLE TITLE PAGE
Table 1: List of Assistive Exoskeletons 5
Table 2: Peak kinetic values during gait cycle 11
Table 3 : The Mechanical properties of the 6061 - T6 Aluminum
Alloy 37
Table 4 : The Physical properties of the 6061 - T6 Aluminum
Alloy 38
LIST OF FIGURES
FIGURE TITLE PAGE
Figure 2.1: The development of lower limb exoskeletons 4
Figure 2.2: Breakdown of the gait cycle into phases based on the
work of Perry and Burnfield (2010) 8
Figure 2.3: Muscle action in the gait cycle 9
Figure 2.4: Ground Reaction Force During the Gait Cycle 10
Figure 2.5: Muscle force expressed as a percentage of the Muscle
Sum. 10
Figure 2.6: Block diagram of the proposed classification of the
control strategies subparts. 12
Figure 3.1: The base design 29
Figure 3.2 a: hip links and joints 30
Figure 3.3 a: knee links and joints 31

x
Figure 3.4 a: Spring loaded ankle 32
Figure 3.5 : Electronic components & battery backpack 34
Figure 3.6 : Custom shoes fitting 35
Figure 3.7 : with shoe/ backpack 36
Figure 3.8 : With thigh/calf support 36
Figure 3.9 : PLA+ 3d printed join 39
Figure 3.10 : bent sheet metal joint 40
Figure 4.1: Multibody walking simulation 43
Figure 4.2: Multibody block connections 44
Figure 4.3: Multibody block connections 45
Figure 4.4: Simulink control sub-system 46
Figure 4.5: Signal processing sub-system 47
Figure 4.6: Hold and shift sub-system 48
Figure 4.7: Input signal processing sub-system 49
Figure 4.8: PID tuning 50
Figure 4.9: power supply 53
Figure 4.10: battery 54
Figure 4.11: Step down 54
Figure 4.12: Limit switch 55
Figure 4.13: knee motor 55
Figure 4.14: hip motor 56
Figure 4.15: Aluminium v-slot links 57
Figure 4.16: Springs 57
Figure 4.17: Sheet metal joints 58
Figure 4.18: Sheet metal joints 59

xi
Figure 4.19: Motor Driver 60
Figure 4.20: incremental rotary encoder 61
Figure 4.21: Raspberry pi 62
Figure 4.22: Circuit connections 64
Figure 4.23: Assembly 65
Figure 4.24: Fitted Assembly 66
Figure 4.25: Close up of assembled spring-loaded ankle joint 67
Figure 4.26: Close up of assembled joint 67
Figure 4.27: Functional group muscle force profiles during
walking stance phase. 72
Figure 4.28: Hip angle, angular velocity, angular acceleration,
moment and power for all analysed
movements. 73
Figure 4.29: Knee angle, angular velocity, angular
acceleration, moment and power for all
analysed movements. 73
Figure 4.30: Ankle angle, angular velocity, angular
acceleration, moment and power for all
analyzed movements. 74

1
CHAPTER 1
INTRODUCTION
1.1 Background
Lower body wearable robots, or lower limb exoskeletons, have developed
rapidly in the past decade. These devices can be separated into three different
categories: assistive exoskeletons, rehabilitation exoskeletons, and augmentation
exoskeletons. With the recent progress in personal care robots, interest in wearable
exoskeletons has been increasing due to the demand for assistive devices. The bulk
of the focus has been on load augmentation for soldiers/workers, assisting trauma
patients, paraplegics, spinal cord injured persons and for rehabilitation purposes.
Barring the military-focused activities, most of the work to date has focused on
medical applications. However, there is a need to shift attention towards the growing
needs of elderly people, that is, by realizing assistive exoskeletons that can help them
just as much as SCI patients to stay independent and maintain a good quality of life.
Therefore, the goal of the project is to provide better mobility and elevate some of the
blood circulation problems (pressure sores) arising from constant wheelchair usage.
Plus, in general decreasing the inaccessibility compared to wheelchairs, greatly
increasing patient independence.
1.2 Problem Statements:
Motion support for the lower limbs of patients suffering from limited motor
function is the issue this project is targeting. Especially Patients who suffer of partial
lower body paralysis due to spine injuries or caused by a stroke or diagnosed with a

2
degenerative disease such as multiple sclerosis that can cause cognitive impairment,
then it is necessary to endow robots with enough knowledge to automatically adapt
to each situation, while achieving a realistic personal use case for the patients.
Research has found that remaining seated for long periods induces health
issues. It is suggested that passive mechanical loading is necessary for maintaining
bone mineral density (BMD), BMD of long-time wheelchair users is statistically
lower than that of individuals who stand with assisting tools.
1. Pressure ulcers : Pressure ulcers are a type of injury that breaks down the skin
and underlying tissue when an area of skin is placed under constant pressure
for certain period causing tissue ischemia, cessation of nutrition and oxygen
supply to the tissues which wheelchair users also suffer from.
2. Ischial tuberosities: Since a high amount of pressure is applied on the seating
surface for long durations. As a result, assisting devices that keep users
standing upright with better manoeuvrability are required.
3. Cost/effect problem: Mainly the biggest benefit would be reducing the costs
compared to some of the commercially available products right now while
maintaining viable ergonomics comparing to wheelchairs as there are no
other viable alternatives currently, people are forced to move to single storey
apartments, install chair lifts and ramps, change fittings because they are too
high or too low and so on. Such changes can impose significant financial
costs on an individual even though the actual cost of the wheelchair or a
mobility scooter can be quite low and attractive. In fact, the overall costs for
effective adoption can be huge. Replacing these traditional wheel-based
mobility solutions by body-fitting exoskeletons becomes attractive because
minimal changes need to be made to homes and lifestyles for staying active
and independent by allowing for the possibility to continue living in one’s
home for as long as possible.
1.3 Aims and Objectives:
The project objectives are:
• Automated Walk gait
the walking support device is focused on patients who suffer of partial lower body
paralysis due to spine injuries or caused by a stroke. This prototype aims to tackle the
issues discussed above by providing Paralyzed patients with an automated walk gait.

3
• Adjustable design
Relative compatibility with different patient sizes and requirement, noting that some
parts will still need to be made to measure for each patient as is the case for medical
wearable equipment.
• Relative ease of disassembly
The ability to be disassembled without the need of difficult to attain tools
• Quality of life
A personalized device that enhances the patient’s abilities and improves his/her
quality of life as much as possible.
• Control strategies
It aims to allow exploration of the different control strategies which will be discussed
later on, like the use of reinforcement learning to achieve a more natural gait, as well
we will be testing the feasibility of model predictive control as well.
• Learning opportunity
It’s a steppingstone for us as mechatronics students to deepen our understanding of
biomechatronic design & robotics control in general.

4
CHAPTER 2
LITERATURE REVIEW
2.1 Current & past products/research prototypes
Figure 2.1: The development of lower limb exoskeletons
As shown in fig. 2.1 lists many of the lower limb exoskeletons that have been
developed. “Making the user superhuman.” Augmentation exoskeleton users are
generally healthy individuals. For healthy users, predefined trajectories are not
necessary. Instead, control algorithms that follow the user’s limb motion, such as
admittance/impedance control or even positive feedback sensitivity amplification
control, are used. Inaccurate but high power/weight ratio actuators, such as series
elastic actuators (SEA) and pneumatic actuators, are more commonly used in this
category.

5
However, Assistive exoskeletons are mostly used by thoracic-level motor-complete
spinal cord injury (SCI) patients [8][11]. Many of these patients permanently lose the
ability to walk and consequently use wheelchairs, noting that the accessibility of the
wheelchair is limited. As shown in table1&2 a list of different Assistive Exoskeleton.
Table 1: List of Assistive Exoskeletons

6
2.2 Gait cycle & Modelling
The typical walk consists of a repeated gait cycle. The cycle itself contains two
phases a stance phase and a swing phase:
Stance phase: Accounts for 60% of the gait cycle. It can be divided into the heel
strike, support, and toe-off phases.
Swing phase: Accounts for 40% of the cycle. It can be divided into the leg lift and
swing phases.
2.2.1 Weight acceptance (0-12%):
The objectives of weight acceptance are to stabilize the limb, absorb shock and
preserve the progression of the body. This phase can be broken down further into
initial contact and loading response. Initial contact consists of the first 3% of the gait
cycle.
In typical gait, the heel strikes the ground and initiates the rotation over the heel to
foot flat to preserve progression. This motion is the first rocker of the gait cycle.

7
Loading response goes from 3-12% of the gait cycle. In this portion, the knee flexes
slightly in order to absorb shock as the foot falls flat on the ground, stabilizing in
advance of single limb support.
2.2.2 Single limb support (12-50%):
Single limb support involves progression of the body over the foot and weight-
bearing stability. The first sub-phase of single limb support is midstance, which is
seen during the 12-31% of the gait cycle. During midstance, the shank rotates
forward over the supporting foot, creating the second rocker motion of the cycle.
This maintains the forward progression of gait. The second stage of single support is
terminal stance which goes from 31-50% of the gait cycle. During terminal stance,
the center of mass advances out in front of the supporting foot. The heel raises of the
ground as you roll onto the ball of the foot, creating the third rocker motion of the
cycle.
2.2.3 Swing phase (50-100%):
The objectives of the swing phase of gait:
1. Foot clearance over the ground.
2. Forward swing of the limb.
3. Preparation of limb for stance.
4. The swing phase can be broken down into 4 sub-phases.
• Pre-swing takes place during 50-62% of the gait cycle. Pre-swing is
the transition phase between stance and swing, in which the foot is pushed and lifted
off of the ground.
• Initial swing goes from 62-75% of the gait cycle. During initial swing,
the hip, knee, and ankle are flexed to begin advancement of the limb forward and
create clearance of the foot over the ground.
• Mid-swing goes from 75-87% of the gait cycle. During mid-swing,
limb advancement continues, and the thigh reaches its peak advancement.
• Terminal swing is the final phase of the gait cycle going from 87-
100% of the cycle. During terminal swing, the final advancement of the shank takes

8
place, and the foot is positioned for initial foot contact to start the next gait cycle.
Figure 2.2: Breakdown of the gait cycle into phases based on the work of Perry and
Burnfield (2010)

9
2.3 Muscular Forces
We reviewed studies done to calculate the average forces acting on the human
body throughout the gait cycle. Musculoskeletal modelling was used to clarify the
details of muscle force generation during walking. Using the kinematic and kinetic
data from ten participants.
Vertical and horizontal peaks and vertical valley of ground reaction forces (GRF),
weight acceptance and push-off rates, and impulse were calculated and compared
across the three experimental conditions.
Figure 2.3: Muscle action in the gait cycle
The study concluded that Different amounts of body weight unloading
promote different outputs of GRF parameters, even with the same mean walk speed.
The only parameter that was similar among the three experimental conditions was
the weight acceptance rate.

10
Figure 2.4: Ground Reaction Force During the Gait Cycle
The early and late force peaks are separated by a valley at 40% of stance phase.
Several functional muscle groups exhibit consistent timing and shape of the force
curve across participants Check appendix (A) Figure 4.28
At five specific gait events: initial contact; peak in the vertical GRF during the
breaking phase of stance; midstance; peak in the vertical GRF during the propulsive
phase of stance; toe-off.
Figure 2.5: Muscle force expressed as a percentage of the Muscle Sum.

11
2.4 Dynamic model
Because an exoskeleton is a wearable device, it could be assumed that the kinematic
data of the patient’s gait are computed through the internal sensors of the robots
[5][13] in addition to the inertial unit sensors placed on the patient’s limbs. However,
this hypothesis is not valid for kinetics data. The robot’s forces are computed by its
dynamical model as shown in table 2.
Table 2: Peak kinetic values during gait cycle

12
2.5 Control strategies
Figure 2.6: Block diagram of the proposed classification of the control strategies
subparts.
The idea of this classification shown in fig. 2.6 is that any controller in the
literature can be represented by a path that joins the used control blocks. The path

13
does not have to start from the high-level layer and may start directly in the mid-
level.[16]
A controller can have several parallel paths if the controller combines several
strategies at the same time, or successively during the gait. Connecting lines show
the common paths identified in the literature. However, it should be noted that the
lack of a line between two blocks does not mean they cannot be related. For instance,
the outcome of the high-level layer, the “operation mode”, could affect any of the
blocks of the middle-level, but it is not connected to them for the sake of
readability.[16]

14
CHAPTER 3
METHODOLOGY
3.1 Trajectory planning
Reference angles and velocities:
Instead of moving independently like humanoid robots, lower limb wearable robots
for those with walking capability must move in conjunction with the human lower
limbs. Thus, the movement capabilities of the wearable robots must at least match
the human movement capabilities. To achieve such a matching, human
anthropometric information (e.g., segment lengths, masses, inertia values) and the
joint capabilities must be identified to deduce the robotic system specifications
[9][11]. For the identification of joint performance dynamometers can be used,
which can determine the relationship between maximum joint moment and
maximum joint speed.
These analyses could be performed with different populations in terms of age
(e.g., students or elderly people) or with populations that have different levels of
physical athleticism (e.g., athletes or non-athletes). However, knowledge of
maximum joint performance values would not provide us with insights regarding
joint requirements throughout daily life. Designing robots based on the human
performance maxima, might overestimate the required specifications, which could
lead to disadvantages such as increased weight and reduced operating time. An
alternative would be to analyse lower limb joint performance during human daily
movements. Humans have developed highly versatile movement skills.
This includes a wide range from minimal movements as in maintaining
balance during quite stance over movements in place when for example lifting
objects, to movements that are used to ambulate such as walking. All movement

15
tasks can be varied in several dimensions, which will change their biomechanical
characteristic. For example, one task could be to design a powered prosthetic ankle
that is able to assist a person with a transtibial amputation during walking. Walking
is determined by the velocity, the slope of the environment, and the shape of the
ground below the foot. Moreover, during locomotion and movements in place,
differences in body weight and additional payload could be considered, as both will
change the human joint effort. A similar scaling effect might exist for movements
that are performed by two legs, compared to movements that completely rely on a
single leg.
With this project we intend to summarize and analyse the hip, knee, and ankle
joint kinematics and kinetics for a broad range of daily essential and sportive
movements. Instead of performing several biomechanical movement experiments on
our own, available data from literature is used for the analysis. We aim to identify the
most demanding movements in terms of maximum absolute joint angular velocity,
maximum absolute joint angular acceleration, maximum absolute joint moment,
maximum absolute joint power, average absolute power, and joint range of motion.
Additionally, this study is used to investigate if there exist differences in the maxima
of locomotion tasks for the stance and the swing phase, as this could allow
alternative mechanical solutions to mimic either of these phases. We expect that the
non-weight bearing swing phase has increased angular velocity and angular
acceleration requirements, whereas the weight bearing stance phase has increased
requirements in moment and power.
Angle speed reference:
The results of the paper analysis [4] can be found for the hip in (Figure 4.29), the
knee in (Figure 4.30), and the ankle in (Figure 4.31). In appendix A
Range of Motion:
Based on paper’s findings for walking movements, it is recommended to consider the
full motion range of hip (17° to −120°) and knee (2° to −144°) for the design of
wearable robotic limbs or human-like robots. This similarly applies to ankle

16
plantarflexion (40°), while an extended range of motion needs to be taken into
account for the ankle dorsiflexion (−38°).
Angular Velocity and Acceleration:
It is necessary to enable the wearable robot to achieve the maximum angular velocity
and acceleration that was found during recovery in this study. For example, while the
hip and knee of a transtibial amputee try to recover from a tripping event, an artificial
powered prosthetic foot with limited angular velocity and acceleration may not be
able to dorsiflex fast enough to provide ground clearance. In order to sufficiently
enable function for daily life, maximum angular velocities of 500°/s for the hip,
550°/s for the knee and 300°/s for the ankle seem appropriate. The angular
acceleration should be 4,400°/s2, 11,200°/s2, and 8,300°/s2 for the hip, knee, and
ankle, respectively. To perform sportive movements, increasing the maximum
velocity is recommended for the knee and the ankle, and increased maximum angular
acceleration is recommended for the hip and ankle.
Moment and Power:
To provide the capabilities for daily life, a maximum moment of 2.4 Nm/kg for the
hip, 1.5 Nm/kg for the knee and 1.9 Nm/kg for the ankle seem appropriate. A
maximum power of 5.8 W/kg, 4.1 W/kg, and 4.3 W/kg appears recommendable for
the hip, knee, and ankle, respectively.
Human Lower Limb Joint Performance Limits:
While this project focuses on the lower limb joint requirements for movements of
daily life, other movements or increased speeds and loads can require increased
capabilities. To investigate kinematic-kinetic relations without being specific to a
movement, researchers have used dynamometers. It has been shown that with
increasing angular velocity, the maximum possible joint moment is reduced, and that
at certain joint angles, the highest moments can be achieved (Anderson et al., 2007).
For young males (non-athletes), the identified maximum isometric extension and
flexion moment for the hip were 2.8 Nm/kg and 1.9 Nm/kg, for the knee were 2.8
Nm/kg and 1.5 Nm/kg, and for the ankle were 1.6 Nm/kg (plantarflexion) and 0.6
Nm/kg (dorsiflexion). While the identified maximum hip moment is not achieved in

17
the analysed daily life movements, the maximum knee moment is achieved, and the
maximum ankle moment found in 4 m/s running is larger than the values achieved
with the dynamometer.
3.2 Inverse Kinematics & Dynamics
3.2.1 Kinematics:
We have used the Robotics System toolbox that developed by Peter Corke for
MatLab. The Robotics System Toolbox provides tools and algorithms for designing,
simulating, testing, and deploying manipulator and mobile robot applications. The
toolbox includes algorithms for collision checking, path planning, trajectory
generation, forward and inverse kinematics, and dynamics using a rigid body tree
representation. It also includes a library of commercially available industrial robot
models that you can import, visualize, simulate, and use with the reference
applications. We have set our model with DH parameters as follows:
After we have applied our model shown in fig. , we get transformation matrix
as follows:
The first row [cos(q1 + q2), -sin(q1 + q2), 0, 4*cos(q1 + q2) + 4*cos(q1)]
The second row [
(4967757600021511*sin(q1 + q2))/81129638414606681695789005144064,
(4967757600021511*cos(q1 + q2))/81129638414606681695789005144064,
-1,
(4967757600021511*sin(q1 + q2))/20282409603651670423947251286016 +
(4967757600021511*sin(q1))/20282409603651670423947251286016]
The third row [ sin(q1 + q2),
cos(q1 + q2),

18
4967757600021511/81129638414606681695789005144064,
4*sin(q1 + q2) + 4*sin(q1)]
The fourth row [ 0, 0, 0, 1]
3.2.2 Dynamics:
The dynamic is described in terms of the time rate of change of the
Exoskeleton configuration, in relation to the joint torques exerted by the actuators.
This relationship can be expressed mathematically as a set of differential equations
known as equations of motion, which describes the dynamic response of the
exoskeleton links to input joint torques. There are two main approaches for obtaining
the dynamic model of any mechanism, the Euler-Lagrange method which is an
energy method, and the Newton-Euler method which is based on the equilibrium of
forces and torques.
The Newton-Euler formulation is derived directly from Newton's Second
Law of Motion, which describes dynamic systems in terms of force and momentum.
The equations combine all the forces and moments acting on the individual robot
links, including the coupling forces and moments between the links. The equations
achieved from the Newton-Euler method include the constraint forces acting between
adjacent links. As a result, additional arithmetic operations are required to remove
these terms and obtain explicit relationships between joint torques and resultant
motion in terms of joint displacements.
In the Euler-Lagrange formulation, on the other hand, the system's dynamic
model is described in terms of work and energy using any set of coordinates, not just
the standard Cartesian coordinates. As a result, all workless and constraint forces are
eliminated automatically in this method. The resulting equations are compact in
general and provide a closed-form expression for joint torques and joint
displacements. Furthermore, unlike the Newton-Euler method, the derivation is
simpler and more systematic.
In our case we need to provide more detailed information about the
Exoskeleton in additional to torque exerted by actuators such as angular velocity,
linear velocity, the force at centre of gravity and the moment of centre of gravity of

19
each link. So, we will go with the recursive Newton-Euler method (inverse
dynamics) and using MATLAB to solve the calculation of matrices.
Before starting our MATLAB program, we need to calculate and set our
parameters. First, we calculate the human body mass and centre of mass from the
distribution of human body masses[3].
Human body mass mb =4.4 + 1.2 + 30.5 + 2.9 + 14.6 + (2.4 + 1.6 + 0.6) *2 + (11.8 +
4.5 + 1.1) *2 = 97.6 Kg.
∑𝑦𝑖𝑚𝑖 = 178*4.4 + 163.8*1.2 + 138.7*30.5 + 117.2*2.9 + 104*14.6 + (151.7*2.4 +
151.7*1.6 + 151.70.6) *2 + (79.8*11.8 + 35.3*4.5 +3*1.1) *2 = 10678.91
The centre of mass of human body from the ground =
∑𝑦𝑖𝑚𝑖
𝛴𝑚𝑖
= 10678.91 / 97.6 =
109.415 cm which is 21.015 cm above the hip pivot.
So, the external force acting on the end effector = 97.6 * 9.8 = 956.48 newton.
And the external moment acting on the end effector about z-axis = 97.6*9.8*0.2075=
198.4696 newton meter.
And the external moment acting on the end effector about y-axis = 97.6*9.8*1.09415
= 1046.5325 newton meter.
The mass of the lower limb of human body mL = mthigh + mcalf + mfoot = 11.8 + 4.5 +
1.1 = 17.4 Kg.
∑𝑦𝑖𝑚𝑖 = 79.8*11.8 + 35.3*4.5 +3*1.1 = 1103.79
The centre of mass of lower limb of human body from the ground =
∑𝑦𝑖𝑚𝑖
𝛴𝑚𝑖
=
1103.79/17.4 = 63.4362 cm.

20
So, the external moment acting on hip joint = 17.4*9.8*0.373638 = 63.7127 newton
meter.
The mass of the calf and foot of human body mc = mcalf + mfoot = 4.5 + 1.1 = 5.6 Kg
∑𝑦𝑖𝑚𝑖 = 35.3*4.5 +3*1.1 = 162.15
The centre of mass of lower limb of human body from the ground =
∑𝑦𝑖𝑚𝑖
𝛴𝑚𝑖
=
162.15/5.6 = 25.9553 cm.
So, the external moment acting on hip joint = 5.6*9.8*0.276447 = 15.1714 newton
meter
The ci
Ii is the inertia tensor of the link i w.r.t. a frame is located at the centre of mass
of the link i and has a same orientation as the link frame. The inertia tensor is given
by:
where the scalar elements are given by:
and

21
The Newton-Euler Equaions that we will run in our code is:
The MatLab code:
First we initialize the DH parameters,

22
Then the masses and the lengths of each link,
Then the values of 𝜃̇ and 𝜃̈ [4]
Then the frame’s location of centre of gravity
Then the initial condions
Then calculating the rotational matrix of each link

23
And calculating the inertia tensor matrix:
After setting the parameters, the for loop start to calculate the forward equations

24
After setting the parameters, the for loop start to calculate the inverse equations
The output will be as folllows:
The angular velocity and angular acceleration
The liner velocity and linear acceleration

25
The linear acceleration of the center of gravity of each link
The inertial force and moment acting at the center of gravity
The forces and moment acting on each link
And finally the tourqe of each joint is

26
3.3 Design
The wearable exoskeleton robots which integrate the human intelligence and
adaptability with the ability and accuracy of robotic system are showing great
potential. The increasingly concerned lower limb exoskeletons are widely used for
walking aid, rehabilitation, load carrying, and body support[1].
In step with their power-assisted effect, they'll be subdivided into:
(1) The follow up type.
(2) The active power-assist type.
The follow-up exoskeletons, which transfer all the load weight to the bottom
(not to the human or wearer). They’re just following the figure motion and that they
are mostly employed in load-carrying operations. On other hand, the active power-
assist-type exoskeletons treat the wearer body as a load and help reduce the energy
consumption of the body by actively assisting the wearer movements.
This feature not just to covers the load-bearing functions of follow-up type
but also expands the movement performance of the wearer, which makes such
exoskeletons quite beneficial for patients’ rehabilitation training, walking aids for the
elderly and disabled persons, and military long-distance travels on difficult ground.
However, it's still being difficult to understand the active power-assist concept and
structural design.
3.3.1 The Design main idea & procedure:
Our Exoskeleton robot is active power-assist that must be as follows: the
applied torque TE exerted on the human body by exoskeleton has the same direction

27
as the joint torque TH, which is required by human body to keep its state of
movement. The joint torque of human body can be reduced, only when TE and TH are
in the same direction. When TE approaches TH, the wearer can move with no effort.
When TE exceeds TH, the exoskeleton is fully controlled the motion. Because the
exoskeleton's applied force is exerted on human skin, which is not as powerful as
bones and ligaments, too much force can lead to discomfort.
Our Exoskeleton robot is active power-assist that must be as follows: the
applied torque TE exerted on the human body by exoskeleton has the identical
direction of the joint torque TH, which is required by the body to stay its state of
movement. The joint torque of human will be reduced, only when TE and TH are
within the same direction. When TE approaches TH, the wearer can move with no
effort[2]. When TE exceeds TH, the exoskeleton is fully controlled the motion. Since
the applied force from exoskeleton is exerted on the human skin, which isn't as
strong as bones and ligaments, an excessive amount of force can cause discomfort.
Based on the above, the focus will be on the following functions of the
mechanical structure:
First is establishing a unique motion mapping between the human body and the
exoskeleton, so that we can acquire human motion data according to the sensing
system in the exoskeleton structure, like a master–slave teleoperation robots.
Specifically, we equivalent the motion intention of the wearer to his or her muscular
torque. But We will work in a different way as we plan to use inverse dynamics to
calculate joint torque of human legs instead of EMG signals.
Second is transmitting assistive force to the wearer. The control objective of follow-
up type exoskeleton is to reduce and even eliminate the man–exoskeleton contact
force. On the contrary, the active power-assist exoskeleton would actively exert the
applied force on the wear body. For ankle assist, force will be added on shank and
foot. For knee assist, force is added on thigh and shank, or directly transfer from the
thigh to the ground, those for weight bearing, just like the exoskeleton robots for

28
patients with spinal cord injury and therefore the walking assist exoskeleton mainly
apply force on the upper thigh and also the torso.
Third is coordinative motion with the wearer. There are two control centres within
the man–exoskeleton coupled system. One is that the human brain, and therefore the
other is that the exoskeleton controller. To confirm the wearer has absolutely the
initiative, the exoskeleton should only exert force on the body, but not forcibly
change the trajectory of human movement. Both the positioning servo and therefore
the speed servo strategy aren't the simplest choice for the control although they're
giving control system feedback, but they need a high stiffness. Another choice is
series elastic actuator and impedance control with position feedback. The previous
has inherent flexibility, but its frequency response is difficult to improve; the last
makes use of other elastomer within the system like the band or the human tissue, but
the accurate stiffness characteristic is difficult to acquire which can cause instability
of the control loop. Therefore, we'll go with dc motor with encoder which may track
the specified position curve rapidly and accurately.
3.3.2 The Design Configuration:
Since the exoskeleton is created to be a wearable device, its configuration, scale, and
number of degrees of freedom (DOFs) should be like those of a skeletal system of
the human. However, the difference within the spatial distribution would inevitably
cause the imperfect agreement between them. Moreover, it's necessary to get rid of
some insignificant elements and use equivalent mechanism transformation to
minimize its complexity. The designed configuration of our exoskeleton is shown in
Figure 3.1.
Hip Joint:
For hip joints, there'll be one rotational DOF adopted to simulate the
flexion/extension function of the human hip joint, as flexion/extension DOF has the
most influence on the foot movement space. The remaining 2 DOFs that complete
spherical hinge of hip has a small motion amplitude specially within the gait cycle,

29
so that they are going to be neglected. To facilitate the structural design, the axis of
this DOF is transferred laterally to the out part of the skin of the human hip as shown
in Figure 3.2 a and Figure 3.2b.
Figure 3.1: The base design

30
3.3.3 Knee Joint:
The human knee joint is a synovial joint, and therefore its kinematics characteristic is
much complicated. However, the general performance mainly corresponds to the
flexion/extension DOF. Therefore, a revolute joint is utilized for the knee joint as
shown in Figure 3.3a and 3.3b. Since this simplified design will cause a relative
motion between the exoskeleton and therefore the human knee joint, it is
inappropriate to attach exoskeleton knee to the wear body.
3.3.4 Ankle Joint:
The ankle joint can even be though as a spherical hinge, which possesses
flexion/extension, ectropion/introversion, and abduction/adduction DOFs. Among
these, the motion range of flexion/extension DOF is that the largest, and this DOF is
that the key to confirm a decent contact between foot and ground.
Figure 3.2 b hip links and joints
Figure 3.2 a: hip links and joints

31
Therefore, it's realized using the rotational DOF with an axis passing through
the spherical center of the human ankle joint. Abduction/adduction is critical to
regulate the movement balance within the left and right directions, which must be
retained. However, since the length of rotating arm of this DOF is brief (about 70 mm),
the transfer of the axis to the surface won't cause significant man exoskeleton relative
movements. The ankle abduction/adduction DOF needs careful consideration because
the function of this DOF is repetitive to it of the ankle joint flexion/extension rotation.
If both DOFs are placed for the exoskeleton, the rotational movement of the
exoskeleton foot are going to be uncertain, which might cause the ambiguous mapping
relationship between human and exoskeleton motions. Therefore, the
abduction/adduction DOF is excluded. Two springs are going to be attached to the
Figure 3.3 b knee links and joints
Figure 3.3 a: knee links and joints

32
exoskeleton ankle joint rather than actuators (passive DOF) as shown in Figure 4a and
Figure 4b.
3.3.5 Man–Exoskeleton coupling:
The man–exoskeleton coupling mechanism is that the physical boundary between the
wearer and the exoskeleton robot. The key point includes DOF matching and force
transmission. The kinematic coupling ensures the entire accordance of the
exoskeleton and wearer body at the connection point. The force transmission
function allows transfer of the assistive force to proper points on wearer body.
The DOF matching ensures that the DOFs of the man–exoskeleton coupling
system are same as those of a human body. An efficient way to reduce load on the
wearer is to line up a force transmitting bypass. Hence, we've to spot the source, the
way transmitted to the bottom, and also the set location of the connection point of the
load. Obviously, the weight is that the source of the load. It consists of gravity,
inertia, and friction. The load produced in each a part of body passes along latter
limbs until to the bottom, causing joint forces and moments.
Figure 3.4 a: Spring loaded ankle Figure 3.4 b: Spring loaded ankle

33
3.3.6 Torso coupling:
For the stance phase, the most load comes from the human torso (including
the arms and the head), and also the cumulative stress within the force transfer point
on human body (including the thigh, shank, spine, pelvis, and foot) gradually
increases [2].
Here, we select torso as a connect point. The pelvis is an interface between
the legs and the torso. It can tolerate heavy loads, and almost no distortion occurs
during movement. Therefore, we select pelvis because the main force transmission
area on the torso. The iliac crest on the sting of the pelvis can bear the vertical force.
The ilium wings, which protrude outward on either side, can bear the lateral forces.
The posterior superior iliac spine and anterior one can bear the reverse and the
forward directed forces, respectively. Meanwhile, since the pelvis width obviously
exceeds its thickness, it can be set to bear the torque round the spine. Thus, the pelvis
can bear the 3-directional forces and one torque moment.
The straps are going to be connected to backpack of the exoskeleton to extend
the connection between the chassis and exoskeleton within the back. After adding the
torso coupling the Exoskeleton are as shown in Figure 3.5.
3.3.7 Foot coupling:
For a swing leg, hip joint is that the base, and foot is that the terminal. So, as
to decrease the load on hip and hinge joint, we'll hold the thigh and therefore the
shank. Also, a connect point on the foot end should be created. The man–exoskeleton
coupling mechanism on the foot enables the synchronous movements. Meanwhile,
human hip and knee joint torque would be minimized within the swing phase stage
since the exoskeleton will lift the wearer foot.
The foot connection is shown in Figure 3.5. As there's a small difference
within the exoskeleton and the joint distributions, and also the exoskeleton
dimensions are adjusted to be same as human body dimensions, the exoskeleton and
human lower limbs are fully consistent within the gait cycle. So, the lower limb
exoskeleton is firmly connected to the human leg, besides a set constraint is applied
to human foot, which will be a shoe that cover all the wearer foot.

34
Figure 3.5 : Electronic components & battery backpack
The exoskeleton shoe is firmly connected to the foot arch with a rigid flat zone at the
rear of the foot. The face of the foot may be flexed to suit the rotation of toe joints.
After adding the foot coupling the Exoskeleton are going to be as shown in Figure
3.6.

35
Figure 3.6 : Custom shoes fitting
3.3.8 Links coupling:
The torso coupling and foot coupling isn’t enough to transmit the force from the
exoskeleton to the human body as there a long distance between them, so we need to
add another connecting point on the links to help transmit the force. We will add two
connecting point on each leg, one on the centre of mass of thigh link and another one
on the centre of mass of calf link as shown in Figure 3.8.
The full Exoskeleton after coupling elements shown in Figure 3.7.

36
Figure 3.8 : With thigh/calf support Figure 3.7 : with shoe/ backpack

37
3.4 Material selection
We have decided the geometric shape of the Exoskeleton and its dimension is
set according to the human body dimension, and we have decided the torque needed
to our exoskeleton prototype. Now we need to select the material used for the
exoskeleton.
3.4.1 Links Material:
As the Exoskeleton robot is a wearable device, it should be light as much as
possible to decrease the load. Therefore, the mechanical properties not only the
important one, we have to into account the physical properties also like the density of
the material. Then, the selected material should be able to bear the load, to be light
and for sure to be cheap. One of the most commercial alloys is 6061 - T6 Extrusions
which is an aluminum alloy. The mechanical properties and physical properties of
the 6061 - T6 Aluminum Alloy is shown in table 3 and table 4:
Table 2 : The Mechanical properties of the 6061 - T6 Aluminum Alloy
Ultimate Tensile Strength 38 ksi (262 MPa)
Max Yield Strength 35 ksi (241 MPa)
Percentage of Elongation 8% (at a thickness of less than 0.25″),
10% (for higher than 0.25″)
Elongation at Break 12% (for a 1/16″ thickness), 17% (for a
thickness of 1/2″).
Brinell Hardness 95
Vickers Hardness 107
Knoop Hardness 120
Rockwell Hardness 40 (scale A), 60 (scale B)
Ultimate Shearing Strength 30 ksi (207 MPa)
Modulus of Elasticity 10000 ksi (68.9 GPa)
Poisson’s Ratio 0.33

38
Fatigue Strength 96.5 MPa
Shear Modulus 3770 ksi (26 GPa)
Shear Strength 30000 ksi (207 MPa)
Table 3 : The Physical properties of the 6061 - T6 Aluminum Alloy
Melting Onset 1080 °F (580 °C)
Thermal Conductivity 170 W/m-K
Thermal Expansion 24 μm/m-K
Specific Heat Capacity 900 J/kg-K
Electrical Conductivity 43% IACS (equal volume), 140% (equal
weight)
Calomel Potential -740 mV
Density 2.7 g/cm3
Embodied Energy 150 MJ/kg
Ultimate Resilience 30 MJ/m3
Modulus of Resilience 520 kJ/m3
Stiffness to Weight 14 (axial), 50 (bending)
Strength to Weight 31 (axial), 37 (bending)
Thermal Diffusivity 68 m2/s
Thermal Shock Resistance 14
These thick extrusions offer high rigidity and are suitable for use in high load.
Its Weldability for Arc: Very Good, Solderability: Good and the Machinability:
Acceptable. Based on the above we have selected the 6061 - T6 Aluminum Alloy
Extrusion as the material for our link.

39
3.4.2 Joint Material:
There are both shear and torsion forces effect on the joint part. The joint part
is a little complex as shown in Figure 3.9, so it could be made by a 3d printing
machine or casting. The filament of 3d printing can’t bear the shear force as its
strength is only 80 MPa, and the casting need a prototype of each part which lead to
increasing the cost. So, we are going to change the geometric shape of the part to be
able to manufacture it of sheet metal. The new shape of the part is as shown in Figure
3.10. The selected sheet metal will be 3 mm thickness to be capable to the torsion
and shear load.
Figure 3.9 : PLA+ 3d printed join

40
3.5 Control strategy
For controlling our model, we used a Raspberry pi 3 model b using its GPIO
with MATLAB Simulink to control it.
MATLAB: It is a proprietary multi-paradigm programming language and numeric
computing environment developed by MathWorks. MATLAB
allows matrix manipulations, plotting of functions and data, implementation
of algorithms, creation of user interfaces, and interfacing with programs written in
other languages[14]. Although MATLAB is intended primarily for numeric
computing, an optional toolbox uses the MuPAD symbolic engine allowing access
to symbolic computing abilities. An additional package, Simulink, adds graphical
multi-domain simulation and model-based design for dynamic and embedded
Figure 3.10 : bent sheet metal joint

41
systems[12]. As of 2020, MATLAB has more than 4 million users worldwide. They
come from various backgrounds of engineering, science, and economics.[6]
Simulink:
Simulink is a MATLAB-based graphical programming environment for modelling,
simulating and analysing multidomain dynamical systems. Its primary interface is
a graphical block diagramming tool and a customizable set of block libraries. It
offers tight integration with the rest of the MATLAB environment and can either
drive MATLAB or be scripted from it. Simulink is widely used in automatic
control and digital signal processing for multidomain simulation and model-based
design.[10]

42
CHAPTER 4
RESULTS AND DISCUSSIONS
4.1 Simulation
For our model we made a simulation on MATLAB Simscape multibody to visualize
our project and test if our calculations are correct.
Multibody:
Simscape Multibody (formerly SimMechanics) provides a multibody simulation
environment for 3D mechanical systems, such as robots, vehicle suspensions,
construction equipment, and aircraft landing gear. You can model multibody systems
using blocks representing bodies, joints, constraints, force elements, and sensors.
Simscape Multibody formulates and solves the equations of motion for the complete
mechanical system. You can import complete CAD assemblies, including all masses,
inertias, joints, constraints, and 3D geometry, into your model. An automatically
generated 3D animation lets you visualize the system dynamics.
Simscape Multibody helps you develop control systems and test system-level
performance. You can parameterize your models using MATLAB variables and
expressions, and design control systems for your multibody system in Simulink. You
can integrate hydraulic, electrical, pneumatic, and other physical systems into your
model using components from the Simscape family of products. To deploy your
models to other simulation environments, including hardware-in-the-loop (HIL)
systems, Simscape Multibody supports C-code generation.

43
4.1.1 Control model /Simscape multibody
First, we created our model on multibody by importing the design model from solid
works and adjust it to fit our specifications in multibody as in figure 4.1. Then we
made a direct motion simulation with our angle reference [4] that have been added
directly to joints block to check for overlaps and the gait cycle as in figure 4.2. After
that model we added the contact with ground to mimic our project as in figure 4.3.
Figure 4.1: Multibody walking simulation

44
Figure 4.2: Multibody block connections

45
Figure 4.3: Multibody block connections

46
4.2 Simulink block diagram
Simulink:
We controlled our motors using encoder for feedback system using MATLAB
Simulink application combined with support package for raspberry pi hardware
library which enables you to create and run Simulink models on Raspberry Pi
hardware.
The support package includes a library of Simulink blocks for configuring
and accessing I/O peripherals and communication interfaces.
It also enables you to interactively monitor and tune algorithms developed in
Simulink as they run on Raspberry Pi. Figure 4.4: Simulink block diagram
Figure 4.4: Simulink control sub-system

47
To synchronize the gait cycle, we added a limit switch at end of each cycle and
programmed the motors to turn in limit switch’s direction until they reach it then stop
and wait for other motors to reach their limits. After all motors reach the starting
point, we then have information about the position of each motor so that we can run
our gait cycle reference synchronized. This is done by using AND logic block for the
gait cycle to start and a direct signal from one limit switch to its corresponding motor
as shown in figure 4.5.
Figure 4.5: Signal processing sub-system
4.2.1 Hold and shift signal:
To start our gait cycle reference in the point where each motor reached its starting
point we needed to hold and shift our signal. This is done by a delay block with
variable length and its length is equal to the time needed to all limit switches to send
its signal. One way to create this variable is by increment its value by the sample
time until it is triggered by the AND block from previous as in figure 4.6.

48
Figure 4.6: Hold and shift sub-system
4.2.2 Signal processing:
To send our signal to the Raspberry pi we had to process it in a way that fit our driver
so one output will control the speed using Pulse-width modulation (PWM) which is a
method of reducing the average power delivered by an electrical signal, by effectively
chopping it up into discrete parts. The average value of voltage fed to the load is
controlled by turning the switch between supply and load on and off at a fast rate. The
longer the switch is on compared to the off periods, the higher the total power supplied
to the load.
Another output will control the direction of rotation. So, (0) output means in forward
direction and (1) output means backward direction. As shown in Figure 4.7 for
reference velocity below (0) the direction output will be (1) meaning that the motor
will move backward. This is done using compare to zero block.

49
Figure 4.7: Input signal processing sub-system
4.2.3 PID control
For a feedback system we used a PID block to control motor response and reduce
steady state error. A proportional integral derivative controller (PID controller) is
a control loop mechanism employing feedback that is widely used in industrial control
systems and a variety of other applications requiring continuously modulated
control.[15]
A PID controller continuously calculates an error value as the difference
between a desired setpoint (SP) and a measured process variable (PV) and applies a
correction based on proportional, integral, and derivative terms (denoted P, I,
and D respectively) as in figure 4.8.

50
Figure 4.8: PID tuning
Simulink support package for Raspberry pi hardware blocks:
4.2.4 PWM block
Generate square waveform on the specified analog output pin. The block input
controls the duty cycle of the square waveform. An input value of 0 produces a 0
percent duty cycle, and an input value of 1.0 produces a 100 percent duty cycle.

51
4.2.5 GPIO write
Sets the logical value of a GPIO pin configured as output.

52
4.2.6 GPIO read
Reads the logical value of a GPIO pin configured as input.
4.2.7 Encoder block
Measure the rotation of a motor

53
4.3 Components Used
4.3.1 Electrical Components
power supply & battery
The main purpose of the power supply was for the testing/R&D phase the
exoskeleton, instead of consuming a lot of batteries figure 4.9.
Figure 4.9: power supply
Voltage: 24 volts
Current: 15 Amp
The battery is mounted for wire-free runs

54
Figure 4.10: battery
Suoer 12V/24V Rechargeable Portable digital Battery Charger (20A)
LM2596 DC-DC buck converter step-down power
This is an LM2596 DC-DC shown in fig. 4.11 is buck converter step-down power
module with a high-precision potentiometer for adjusting output voltage, capable of
driving a load up to 3A with high efficiency. When the output current required is
greater than 2.5A(10W) as in figure 4.11.
Figure 4.11: Step down

55
Lever limit switch
Figure 4.12: Limit switch
- Limit switches shown in fig. 4.12 are used to detect the start of the gait cycle
(reset starting position)
- Safety feature to avoid injuries
Knee motor
-DC motor.
Figure 4.13: knee motor
• Motor diameter/gear length/54/160
• The drive is transmitted by means of a truncated fi 8mm axis, which must be
inserted into the through hole in the gearbox
• Total weight about 1,5kg
• Turnovers - 55obr/min
• No-load current 0.25A with 3A load
• Metal worm gear
• Mounting with three holes in the M6 threaded gearbox

56
Hip motor
Figure 4.14: hip motor

57
4.3.2 Mechanical Components
V slotted Aluminum Rods as links
Figure 4.15: Aluminium v-slot links
Spring for ankle joints
Figure 4.16: Springs

58
Sheet metal as joints
Figure 4.17: Sheet metal joints

59
Figure 4.18: Sheet metal joints

60
4.3.3 Electronic Components
Cytron Dual Channel 10A DC Motor Driver
The 10A 5-30V Dual Channel DC Motor Driver is the dual-channel version of
MD10C. This is a dual motor driver which is designed to drive 2 DC motor with high
current up to 10A continuously by using Raspberry Pi as the controller. and 30A
peak. The driver works in voltage range from 5 to 25 V. It also includes fast test
switch for driver testing.
This driver supports locked-antiphase and sign-magnitude PWM signal. It uses full
solid-state components which result in faster response time and eliminate the wear
and tear of the mechanical relay.
Figure 4.19: Motor Driver
MDD10A has been designed with the capabilities and features of:
• Bi-directional control for 2 brushed DC motors.
• Support motor voltage ranges from 5V to 25V 30V (Rev2.0).
• Maximum current up to 10A continuous and 30A peak (10 second) for each
channel.

61
• Solid state components provide faster response time and eliminate the wear
and tear of mechanical relay.
• Fully NMOS H-Bridge for better efficiency and no heat sink is required.
• Speed control PWM frequency up to 20KHz (Actual output frequency is
same as input frequency).
• Support both locked-antiphase and sign-magnitude PWM operation.
• Support TTL PWM from microcontroller, not PWM from RC receiver.
• Onboard push button to control the motor manually
Omron Rotary Encoder E6B2-CW6C (2000 P/R)
This is Omron quadrature incremental rotary encoder E6B2-CW6C with 2000 P/R
(pulse per revolution). The encoder outputs gray code which you can interpret using
a microcontroller or Arduino and find out which direction the shaft is turning and by
how much.
This allows you to add feedback to motor control systems.
This encoder is of improved reliability with reverse connection and load short-circuit
protection.
Figure 4.20: incremental rotary encoder
Features:
• Resolution: 2000 Pulse/Rotation

62
• Encoding Method: Incremental
• Input Voltage: 5 - 24VDC
• Maximum Rotating Speed: 6000rpm
• Allowable Radial Load: 30N
• Allowable Axial Load: 20N
Raspberry Pi 3B:
Figure 4.21: Raspberry pi
Raspberry Pi is a series of small, single-board computers, the Raspberry Pi can
interact with the outside world. The operating system for all Raspberry Pi products is
Linux. Linux is an open-source operating system that interfaces between the
computer’s hardware and software programs. The language used with Raspberry Pi
is Python. a general-purpose and high-level programming language used to develop
graphical user interface (GUI) applications, websites, and web applications. One of
the benefits of Raspberry Pi is that it is not necessary to have an intimate knowledge
of Linux or Python before beginning a project with Raspberry Pi. In fact, the purpose
of the product is to teach the system and language through engaging projects. It runs
Linux (a computer operating system) and provides general-purpose input and output
(GPIO) pins that allow the user to control electronic components for physical

63
computing and exploring the Internet of Things (IoT). With the Raspberry Pi 3B
relatively new, it has many advantages over other microcontrollers and
microcomputer (ex: Arduino UNO).
Raspberry pi 3 model B:
Raspberry pi 3 model B is the model of the third-generation Raspberry Pi. It replaced
Raspberry Pi 2 Model B in February 2016. Its specifications are:
• Quad Core 1.2GHz Broadcom BCM2837 64bit CPU
• 1GB RAM
• BCM43438 wireless LAN and Bluetooth Low Energy (BLE) on board
• 100 Base Ethernet
• 40-pin extended GPIO
• 4 USB 2 ports
• 4 Pole stereo output and composite video port
• Full size HDMI
• CSI camera port for connecting a Raspberry Pi camera
• DSI display port for connecting a Raspberry Pi touchscreen display
• Micro SD port for loading your operating system and storing data
• Upgraded switched Micro USB power source up to 2.5A

64
4.4 Circuit connections
The following is an illustration of the circuit connections for our prototype
Figure 4.22: Circuit connections

65
4.5 Assembly
Noting the following images isn’t the final form of the prototype, it’s still missing the
thigh & calf supports, additionally the elastic straps and proper routing of the wires.
Figure 4.23: Assembly

66
Figure 4.24: Fitted Assembly

67
Figure 4.26: Close up of assembled joint
Figure 4.25: Close up of assembled spring-loaded ankle joint

68
CONCLUSION AND RECOMMENDATIONS
4.6 Limitations
This project is definitely highly complex and requires a good grasp of multiple
disciplines Bioengineering, Mechanical, electronics, robotics & control theory etc.
Which is one of the reasons we chose it in the first place, in order to have the
chance to dive in those fields and improve our knowledge and practical experience.
However, it’s advised to include bioengineering expertise in the improvement of this
prototype, to further focus on what matters, which is how can the device affect the
patient’s life and improve, while still maintains realistic use scenario.
One of the limitations faced is attaining the proper motors for actual use on humans
(able to hold the max torque requirements on the joints).
4.7 Future work/recommendations
• While geared DC brushless motor is the lightest & best option that can hit the
requirements, Recommendation is importing said motors in a timely manner
if the situation allows. DC worm geared motor is recommended for lower
cost, there is multiple lighter options than the DC motor that was available to
us for use on the knee.
• The humanoid and exoskeleton system should be simulated with different
conditions such as disturbance forces, constant loads, and different motion
cycle speeds, to evaluate the control efficiency and robustness.
• A humanoid model should be designed in a way to replicate dynamics and
kinematics of the human body.
• Testing out different control strategies on the prototype itself for robustness
and effectiveness.

69
• A proper intention [7] of motion detection system which is automated via AI
to make the motion more natural (starting and stopping). As well as a
detection system for mode selecting (stair climbing – standing/sitting – rough
terrain).
4.8 Conclusion
A comprehensive literature review has been carried out to gain understanding
of what has been achieved in the exoskeleton technology there are no sufficient
experimental publications that provide medical and technical assessment for using
exoskeleton devices specially fo r standing up and sitting down movements.
This prototype proves the possibility for a simple low-cost design for testing
the different control strategies/ automation of initiation of the gait cycle/ detecting
intention. It has been adequate to provide a platform to test the controllers and
simulate the designated mobility task, even the viability for different use cases beside
assisting. The presented results obtained through simulation of the humanoid and
exoskeleton system may be considered for further analysis and development.
4.9 Closing words
Wearable robots have long been dreamed of in science fiction. They are often
described as a mixture of robot and clothing: tools and users are no longer separated.
This idea has been realized both in industry and academia in the past decade.
It’s time to put some work into this untapped potential in Egypt

70
REFERENCES
1. Minchala, L. I., Astudillo‐Salinas, F., Palacio‐Baus, K., & Vazquez‐Rodas,
A. (2017). Mechatronic design of a lower limb exoskeleton. Design, Control
and Applications of Mechatronic Systems in Engineering.
URL: https://doi.org/10.5772/67460
2. Deng, J., Wang, P., Li, M., Guo, W., Zha, F., & Wang, X. (2017). Structure
design of active power-assist lower limb exoskeleton Apal Robot. Advances
in Mechanical Engineering, 9(11), 168781401773579.
URL: https://doi.org/10.1177/1687814017735791
3. Anthropology Research Project. (1988). Anthropometry and mass
distribution for human analogues.
URL: https://www.humanics-es.com/ADA304353.pdf
4. Martin Grimmer, Ahmed A. Elshamanhory and Philipp Beckerle, 2020.
“Human Lower Limb Joint Biomechanics in Daily Life Activities: A
Literature Based Requirement Analysis for Anthropomorphic Robot
Design”, Frontiers in Robotics and AI
URL: https://doi.org/10.3389/frobt.2020.00013
5. Jesús Tamez-Duque, 2021. “Alice Open-Source Exoskeleton”, hackaday
URL: https://hackaday.io/project/176681-alice-open-source-exoskeleton-
2021-update
6. Company overview - mathworks. (n.d.). Retrieved July 13, 2022.
URL: https://www.mathworks.com/content/dam/mathworks/fact-
sheet/company-fact-sheet-8282v19.pdf
7. Laschowski, B., McNally, W., Wong, A. and John McPhee 2021. “Computer
Vision and Deep Learning for Environment-Adaptive Control of Robotic
Lower-Limb Exoskeletons”
URL: https://www.biorxiv.org/content/10.1101/2021.04.02.438126v1
8. Baltej Singh Rupal1, Sajid Rafique, Ashish Singla, Ekta Singla, Magnus
Isaksson and Gurvinder Singh Virk., 2017 “Lower-limb exoskeletons:
Research trends and regulatory guidelines in medical and non-medical
applications”
URL:https://journals.sagepub.com/doi/pdf/10.1177/1729881417743554

71
9. Jianhua Chen, Xihui Mu, Fengpo Du. 2017 “Biomechanics analysis of
human lower limb during walking for exoskeleton design” Journal of
Vibroengineering, Vol. 19, Issue 7, p. 5527-5539
URL: https://www.extrica.com/article/18459
10. "The Successful development process with MATLAB Simulink in the
framework of ESA's ATV project" (PDF). Vega Group PLC. Archived
from the original (PDF) on 2011-07-17. Retrieved 2011-11-01.
11. Suin Kim, Kyongkwan Ro and Joonbum Bae., 2017 “Estimation of
Individual Muscular Forces of the Lower Limb during Walking Using a
Wearable Sensor System” Journal of Sensors
URL: https://doi.org/10.1155/2017/6747921
12. Baud, R., Manzoori, A.R., Ijspeert, A., 2021“Review of control strategies for
lower-limb exoskeletons to assist gait.” J NeuroEngineering Rehabil 18, 119
URL: https://doi.org/10.1186/s12984-021-00906-3
13. Manuel Cardona, Cecilia E., García Cena, Fernando Serran and Roque
Saltaren 2020 “ALICE: Conceptual Development of a Lower Limb
Exoskeleton Robot Driven by an On-Board Musculoskeletal Simulator”
Sensors, 20(3), 789
14. Wikimedia Foundation. (2022, July 12). Matlab. Wikipedia. Retrieved July
13, 2022.
URL: https://en.wikipedia.org/wiki/MATLAB
15. Wikimedia Foundation. (2022, July 8). Pid Controller. Wikipedia. Retrieved
July 13, 2022.
URL: https://en.wikipedia.org/wiki/PID_controller
16. Hao Lee, Peter Walker Ferguson, and Jacob Rosen 2020 “Lower Limb
Exoskeleton Systems—Overview” Sensors, p 207-229

72
APPENDICES
APPENDIX A: Graphs
Figure 4.27: Functional group muscle force profiles during walking stance phase.

73
Figure 4.28: Hip angle, angular velocity, angular acceleration, moment and power for
all analysed movements.
Figure 4.29: Knee angle, angular velocity, angular acceleration, moment and power for
all analysed movements.

74
Figure 4.30: Ankle angle, angular velocity, angular acceleration, moment and power for
all analyzed movements.

Lower limp exoskeleton

Recommended

Recommended

More Related Content

Similar to Lower limp exoskeleton

Similar to Lower limp exoskeleton (20)

Recently uploaded

Recently uploaded (20)

Lower limp exoskeleton