load cell for optimizze uad fo r the ahab

1
Human-
Exoskeleton
Interaction

2
Human -Exoskeleton Interaction-Electronics and
Software

.
Methodology-Exoskeleton Kinematics and Force
: The exoskeleton connected to the human arm. If the exoskeleton applies a
torque, this torque is added to the human’s torque
The exoskeleton mechanics: the motor at the joint creates a torque Texo that
is transferred through the mechanical interfaces to the human inside

Methodology-Electronics and Software

Electronic circuit assembly:
8

• The provided Arduino code is designed to interface with an HX711
load cell amplifier and a servo motor to measure weight and control
the servo position based on the detected weight.
10
Arduino code:

Control
Systems
• we will develop and implement different control systems that define the
exoskeleton’s behavior.
• To do so, we will define a set of rules of how the EXOSKELETON should
act and react when worn by the human, pack those rules into some
equations, and implement them on our Arduino microcontroller.
11
Control diagram of a generic control loop

Control Systems
• On the left is the desired value of a variable or system state, such as position, force, or
velocity.
• This is compared to the values measured by a sensor (bottom of diagram, feedback
loop), and the difference is calculated. This difference is then sent to the controller
that adjusts, for example, the motor current, opens and closes a valve, or executes an
action to ’push’ the system (called ’plant’) towards the desired value.
• After this input goes through the plant, the sensors measure again if the desired value
is achieved, and the controller reacts accordingly.
• This is repeated over and over again to put and keep the system at the desired value.
The desired value can change over time, and the job of the controller is to make the
system follow these changes
12

Example: Control of
a Gait Restoration
Exoskeleton
• To control an exoskeleton
for gait restoration in users
with paraplegia, manual
control inputs combined
with a position controller
are used. As the user’s legs
are completely paralyzed,
the exoskeleton has to
provide all the power
required to move them
13

Example: Control of
a Gait Restoration
Exoskeleton
• In this case, the high-level control is
implemented as manual control by the user.
• Buttons located in the handles of crutches
(that are required anyway for balancing) are
used to select and initiate movements
manually.
• Different types of movements are
prerecorded and stored in the exoskeleton’s
control systems.
• These recordings provide the desired angle
trajectories for each of the actuated
exoskeleton joints.
• When a movement is triggered by the user,
the low-level joint position controllers
execute the recorded joint angles, replaying
the movement. Then, the exoskeleton
automatically continues to execute
consecutive steps until stopped by the user,
or each step can be triggered manually by the
user, who then remains in command
14

Example: Control
of a Work Assist
Exoskeleton
• Upper-body exoskeleton that supports the user when lifting
heavy objects. The device is connected to the user’s back
and hip like a backpack, and relieves the user’s arms when
lifting and carrying a payload
• To detect when the user needs support, sensorized gloves
are used in the high-level control. A pressure sensor in each
palm measures when the exoskeleton user is carrying a
heavy object, and only then a desired vertical assistive
force is activated.
• The highlevel controller calculates which joint torques are
necessary in each arm joint to provide the desired support.
These desired joint torques are then sent to the low-level
torque controllers of the joints, and they take it from there
15

18
Gait Analysis Methods
Data collection techniques: Motion capture systems:
Capturing 3D movement using markers and cameras.
Force sensors: Measuring ground reaction forces and joint forces
during walking.
Biomechanical analysis: Joint angles: Analyzing angles at hip, knee,
and ankle joints.
Muscle activation patterns: Assessing muscle activation levels using
electromyography (EMG).
Joint forces and torques: Evaluating forces and torques exerted at
different joints.
Overall analysis: Integrating multiple data sources for a
comprehensive understanding of gait mechanics.

19
Integration and Testing
• Prototype Development: Develop prototypes incorporating the selected actuators and
control systems, ensuring they are integrated seamlessly with the exoskeleton structure.
• Experimental Validation: Perform experimental validation through lab tests and user
trials to verify that the exoskeleton can handle shock and vibratory loads effectively and
provide the necessary support and stability.
• Iterative Refinement: Use feedback from testing to iteratively refine the design,
improving stability, comfort, and overall performance.

20
Kinematic and Dynamic Analysis of the
Exoskeleton

21
Kinematic Analysis
 Kinematic analysis deals with the study of motion without considering the forces causing that motion.
 In the context of exoskeletons, kinematic analysis involves understanding the joint angles, positions, velocities,
accelerations, and trajectories of both the wearer's limbs and the exoskeleton's mechanical components.
 This analysis helps ensure proper alignment, comfort, and safety for the wearer.
Key points in kinematic analysis include:
Joint Angles: Study of joint angles to ensure alignment between wearer's
natural movement and exoskeleton's range of motion.
End-Effector Path: Tracking the path of exoskeleton's end-effectors to evaluate
trajectory accuracy and movement precision.
Workspace Analysis: Determining the exoskeleton's reachable workspace to
assess versatility and suitability for different tasks.
Inverse Kinematics: Solving inverse kinematics problems to control actuators
and achieve desired end-effector positions.

22
Kinematic Modeling
Forward Kinematics describes the position of the end-effector (or tip) of the robot in terms of its joint angles.
• For a 2-DOF planar robot:
•Where:
• are the joint angles.
• are the lengths of the first and second link, respectively.
Jacobian relates the joint velocities to the linear and angular velocities of the end-effector. For our 2-DOF
planar robot, the Jacobian is a 2x2 matrix:

23
Kinematic Modelling
• The exoskeleton can be simplified into a five-link mechanism.
• Support phase and swing phase during walking are modelled.
• Only the motion in the sagittal plane is considered.
• The supporting leg can be equivalent to a two-link system in which
the foot is connected with the geodetic coordinate system.
• The swinging leg can be equivalent to a two-link system connected
with the waist coordinate system.

25
Five Link Model
• -> Motion angle for ankle joint
• and -> Motion angles for knee joints
• and -> Motion angles for hip joints
• and -> Distance between ankle joints and knee joints
• and -> Distance between knee joints and hip joints

26
Denavit–Hartenberg (D-H) method
• The Denavit–Hartenberg (D-H) method is commonly applied in kinematic research of
multijoint structure.
• The four parameters of D-H conventions are Joint Angle , Joint distance Link Length Link twist
angle .
• With those four parameters, we can translate the coordinates from to .

27
D-H parameters
According to D-H method, the z-axis is the direction along the rotation axis
of the joint, and the x axis is the direction of the common normal of the
adjacent z axes.

28
Kinematic model
Assuming that the torso part is vertical so:
The coordinate system fixedly connected to the connecting rod is called coordinate system . Select the
coordinate system fixed on the base as the basic coordinate system. According to the principle of
coordinate transformation, it can be seen that:
, , , ,

29
MATLAB Code
% Symbolic variables
syms q1 q2 q3 q4 q5 l1 l2 l3 l4
% Define transformations
T01 = [cos(q1) 0 -sin(q1) 0; 0 1 0 0; sin(q1) 0 cos(q1) 0; 0 0 0 1];
T12 = [cos(q2) 0 -sin(q2) l1*cos(q2); 0 1 0 0; sin(q2) 0 cos(q2) l1*sin(q2); 0 0 0 1];
T23 = [cos(q3) 0 -sin(q3) l2*cos(q3); 0 1 0 0; sin(q3) 0 cos(q3) l2*sin(q3); 0 0 0 1];
T34= [cos(q4) 0 sin(q4) 0; 0 1 0 0; -sin(q4) 0 cos(q4) 0; 0 0 0 1];
T45 = [cos(q5) -sin(q5) 0 l3*cos(q5); 0 1 0 0; sin(q5) cos(q5) 0 l3*sin(q5); 0 0 0 1];
T56 = [1 0 0 l4; 0 1 0 0; 0 0 1 0; 0 0 0 1];
% Calculate forward kinematics
T03 = T01 * T12 * T23
T36 = T34 * T45 * T56
OUTPUT

30
• Therefore the end position coordinates of the support leg and swing leg
can be obtained:
• Assume that ,

31
Kinematic Modelling – MATLAB Code
% Kinematic Parameters
L1 = 0.5; % Length of first segment (meters)
L2 = 0.4; % Length of second segment (meters)
L3 = 0.5; % Length of third segment (meters)
L4 = 0.4; % Length of fourth segment (meters)
% Joint angles (thetas in radians)
theta2 = pi/4;
theta3 = pi/6;
theta1 = theta2+theta3;
% Kinematic Equations
xh = L2 * cos(theta2+theta3) + L1 * cos(theta3);
zh = L2 * sin(theta2+theta3) + L1 * sin(theta3);
xa = L3 * cos(theta2+theta3) + L4 * cos(theta3);
za = L3 * sin(theta2+theta3) + L4 * sin(theta3);
X=atan(zh/xh);
Y=atan(za/xa);
% Display end-effector position
fprintf('End-Effector Position: (%.2f, %.2f)n', X, Y);
OUTPUT

32
Dynamic Analysis
 Dynamic analysis involves the study of forces, torques, and energy during movement.
 In the context of exoskeletons, dynamic analysis provides insights into the interaction between the
exoskeleton and the wearer's body, as well as the energy requirements for various activities.
 Key points in dynamic analysis include:
Joint Torques: Understanding the torques exerted on each joint helps in designing
actuators that can provide adequate assistance or resistance.
Interaction Forces: Analyzing the forces between the exoskeleton and the
wearer's body is crucial for ensuring user comfort and preventing discomfort or
injury.
Energy Consumption: Evaluating the energy consumption of the exoskeleton
during different activities helps optimize its design for prolonged use and
portability.
Gait Analysis: Studying the dynamics of walking and other movements while
wearing the exoskeleton can provide insights into its effect on natural human
movement patterns.

33
Dynamic model
An exoskeleton is expected to help humans carry heavy loads and
perform demanding tasks for longer periods of time, or it can help
humans save energy in some task.
Dynamic behavior is one of key issues in exoskeleton system.
An exoskeleton with good dynamic behavior can help increase
stability, safety and minimize human-exoskeleton interaction forces.

34
Dynamic Modeling
The Lagrangian method is used to derive the equations of motion.
The Lagrangian is defined as the difference between the kinetic energy and
the potential energy:
Kinetic Energy for each link:
For the first link:
For the second link:
•Where:
• are the masses of the first and second links, respectively.
• are the angular velocities of the joints.

35
Potential Energy for each link:
For the first link:
For the second link:
where is the acceleration due to gravity.
Equations of Motion are derived from the Euler-Lagrange equation:
Where represents the generalized coordinates, which in this case are the joint angles and .
The resulting equations of motion describe the dynamic behaviour of the robot and can be used to
determine the required joint torques for a given motion or external force.

36
MATLAB Code
% Kinetic energy
KE1 = 0.5 * m1 * transpose(v1) * v1 + 0.5 * transpose(w1) * I1 * w1;
KE2 = 0.5 * m2 * transpose(v2) * v2 + 0.5 * transpose(w2) * I2 * w2;
KE = simplify(KE1 + KE2);
% Potential energy
PE1 = m1 * g * lc1 * cos(q1);
PE2 = m2 * g * (l1 * cos(q1) + lc2 * cos(q1 + q2));
PE = simplify(PE1 + PE2);
% Lagrangian
L = KE - PE;
% Euler-Lagrange equation
eq1 = simplify(diff(diff(L, dq1), q1) - diff(L, q1) + ddq1);
eq2 = simplify(diff(diff(L, dq2), q2) - diff(L, q2) + ddq2);
% Solve for second-order accelerations (ddq1 and ddq2)
solutions = solve([eq1 == 0, eq2 == 0], [ddq1, ddq2]);
ddq1_solution = solutions.ddq1
ddq2_solution = solutions.ddq2
OUTPUT

37
Fuzzy Logic
Definition:
 Fuzzy logic is a mathematical framework that deals with reasoning and decision-
making under uncertainty.
 It allows for the representation and manipulation of imprecise or uncertain
information.
Application in Exoskeleton Research:
 Fuzzy logic can be used to develop control algorithms for exoskeletons that can adapt
to varying conditions and user preferences.
 It enables the incorporation of subjective inputs and expert knowledge into the
control system, enhancing the system's flexibility and performance.

38
Fuzzy Rules:
Fuzzy logic provides a way to model complex systems where uncertainty or vagueness is present.
It's based on "degrees of truth" rather than the usual "true or false" (1 or 0) boolean logic.
•In the code:
• A fuzzy inference system is defined to adjust the desired joint torques based on the current joint
angles.
• Membership functions are defined for the inputs (joint angles) and outputs (joint torques).
These functions define how each point in the input space is mapped to a membership value
between 0 and 1.
• Fuzzy rules are then defined to map the input space to the output space. These rules are the
core of the fuzzy logic system and determine the system's behavior.
•

39
Fuzzy Rules – Mathematical Representation
Mathematically, the fuzzy system can be represented as a set of IF-THEN rules:
• 𝐼𝐹 ( ) ( )
𝑥 𝑖𝑠 𝐴 𝑇𝐻𝐸𝑁 𝑦 𝑖𝑠 𝐵
• where:
• 𝑥 is the input variable.
• 𝐴 is the fuzzy set for the input.
• 𝑦 is the output variable.
• 𝐵 is the fuzzy set for the output.
The output is then defuzzified to obtain a crisp value, which can be used for control purposes.
𝑦
•
In MATLAB code first models the dynamics of a 2-DOF exoskeleton robot using the Euler-Lagrange equation.
It then defines a fuzzy logic system to adjust the joint torques based on the current joint angles. The
combination of these two approaches allows for the modelling and control of the exoskeleton robot in the
presence of uncertainties.

40
Neural Networks
Definition:
 Neural networks are computational models inspired by the
structure and function of the human brain.
 They consist of interconnected nodes (neurons) that process and
transmit information.
Application in Gait Analysis:
 Neural networks can analyze gait data to classify gait
abnormalities, detect patterns, and predict future gait patterns.
 They can learn from large datasets, improving their ability to
recognize complex gait patterns and provide accurate
assessments.

41
Fuzzy Neural Networks
Fuzzy Neural Network (FNN) FNN is an AI technique developed from the
grouping of two fields, fuzzy logic and neural network.
FNN detects parameters of a fuzzy system, including fuzzy sets and fuzzy rules, by
manipulating the approximation techniques from neural networks.

42
Artificial Intelligence (AI)
Definition:
 Artificial intelligence (AI) refers to the development of computer systems that can
perform tasks that typically require human intelligence.
 It encompasses various techniques, including machine learning, natural language
processing, and computer vision.
Applications in Exoskeleton Research and Gait Analysis:
 AI can be used to optimize exoskeleton control algorithms based on real-time sensor
data and user feedback.
 It enables adaptive control strategies that can adjust the exoskeleton's behavior based
on the user's intention, gait characteristics, and environmental conditions.

load cell for optimizze uad fo r the ahab

More Related Content

Similar to load cell for optimizze uad fo r the ahab (20)

Recently uploaded (20)

load cell for optimizze uad fo r the ahab