Professional Documents
Culture Documents
Download pdf Biomedical Engineering And Computational Intelligence Proceedings Of The World Thematic Conference Biomedical Engineering And Computational Intelligence Biocom 2018 Joao Manuel R S Tavares ebook full chapter
Download pdf Biomedical Engineering And Computational Intelligence Proceedings Of The World Thematic Conference Biomedical Engineering And Computational Intelligence Biocom 2018 Joao Manuel R S Tavares ebook full chapter
Computational Intelligence:
Proceedings of The World Thematic
Conference—Biomedical Engineering
and Computational Intelligence,
BIOCOM 2018 João Manuel R. S.
Tavares
Visit to download the full and correct content document:
https://textbookfull.com/product/biomedical-engineering-and-computational-intelligenc
e-proceedings-of-the-world-thematic-conference-biomedical-engineering-and-comput
ational-intelligence-biocom-2018-joao-manuel-r-s-tavares/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
https://textbookfull.com/product/proceedings-of-international-
conference-on-computational-intelligence-and-data-engineering-
proceedings-of-iccide-2018-nabendu-chaki/
https://textbookfull.com/product/advances-in-computational-
intelligence-proceedings-of-second-international-conference-on-
computational-intelligence-2018-sudip-kumar-sahana/
https://textbookfull.com/product/proceedings-of-international-
conference-on-computational-intelligence-and-data-engineering-
iccide-2017-1st-edition-nabendu-chaki/
https://textbookfull.com/product/feature-engineering-and-
computational-intelligence-in-ecg-monitoring-chengyu-liu/
Computational Intelligence and Optimization Methods for
Control Engineering Maude Josée Blondin
https://textbookfull.com/product/computational-intelligence-and-
optimization-methods-for-control-engineering-maude-josee-blondin/
https://textbookfull.com/product/sentiment-analysis-and-ontology-
engineering-an-environment-of-computational-intelligence-1st-
edition-witold-pedrycz/
Biomedical Engineering
and Computational
Intelligence
Proceedings of The World Thematic
Conference—Biomedical Engineering
and Computational Intelligence,
BIOCOM 2018
Lecture Notes in Computational Vision
and Biomechanics
Volume 32
Series Editors
João Manuel R. S. Tavares , Departamento de Engenharia Mecânica,
Faculdade de Engenharia, Universidade do Porto, Porto, Portugal
Renato Natal Jorge, Faculdade de Engenharia, Universidade do Porto, Porto,
Portugal
Research related to the analysis of living structures (Biomechanics) has been carried out
extensively in several distinct areas of science, such as, for example, mathematics, mechanical,
physics, informatics, medicine and sports. However, for its successful achievement, numerous
research topics should be considered, such as image processing and analysis, geometric and
numerical modelling, biomechanics, experimental analysis, mechanobiology and Enhanced
visualization, and their application on real cases must be developed and more investigation is
needed. Additionally, enhanced hardware solutions and less invasive devices are demanded. On
the other hand, Image Analysis (Computational Vision) aims to extract a high level of information
from static images or dynamical image sequences. An example of applications involving Image
Analysis can be found in the study of the motion of structures from image sequences, shape
reconstruction from images and medical diagnosis. As a multidisciplinary area, Computational
Vision considers techniques and methods from other disciplines, like from Artificial Intelligence,
Signal Processing, mathematics, physics and informatics. Despite the work that has been done in
this area, more robust and efficient methods of Computational Imaging are still demanded in many
application domains, such as in medicine, and their validation in real scenarios needs to be
examined urgently. Recently, these two branches of science have been increasingly seen as being
strongly connected and related, but no book series or journal has contemplated this increasingly
strong association. Hence, the main goal of this book series in Computational Vision and
Biomechanics (LNCV&B) consists in the provision of a comprehensive forum for discussion on
the current state-of-the-art in these fields by emphasizing their connection. The book series covers
(but is not limited to):
In order to match the scope of the Book Series, each book has to include contents relating, or
combining both Image Analysis and mechanics. Indexed in SCOPUS, Google Scholar and
SpringerLink.
Editors
Biomedical Engineering
and Computational
Intelligence
Proceedings of The World Thematic
Conference—Biomedical Engineering
and Computational Intelligence,
BIOCOM 2018
123
Editors
João Manuel R. S. Tavares Nilanjan Dey
University of Porto Techno India College of Engineering
Porto, Portugal Kolkata, India
Amit Joshi
Global Knowledge Research Foundation
Ahmedabad, Gujarat, India
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Organizational Committees
Conference Chairs
Program Secretary
v
vi Organizational Committees
Advisory Committee
xi
Contents
xiii
xiv Contents
1 Introduction
Human arm has the advantage of smooth dynamic path in its configuration space
and thus can reach several orientations and positions. Such a mechanism encounters
redundancy and thus has more degree of freedoms than needed to operate, which
enrich its smooth motion in difficult areas of the workspace. Studying the relationship
between the joints of a manipulator and the position and rotation of the gripper is
part of the discipline of kinematics [1]. For instance, inverse kinematics (IK) is
essential for controlling the gripper’s location in the workspace by computing joint
variables. However, IK introduces numerous solutions which must be regularized
[2, 3]. Solutions to inverse kinematic problem (IKP) are treated through analytical
approach and the iterative approach or a combination of both techniques. The iterative
approach requires time cost and power consumption. Moreover, it does not always
achieve convergence and may get stuck in an infinite loop. On the other hand, the
analytical method requires less time and power consumption, yet it is only applied for
specific mechanisms. As a result, both techniques are not very consistent in solving
inverse kinematics problem.
A new technique will be developed by considering the arm as a system that consists
of a base made up from a shoulder, upper arm, lower arm, and the hand as the end
effector. This system requires instantaneous joint stiffness as an input parameter
in different movements and positions. Since muscles in our system are the active
components (actuators), the stiffness in the joints is driven by muscle stiffness.
The IKP is the problem in finding a set of joint variables to produce desired end-
effector position and orientation. In such a case, two situations exist. The first is when
many solutions exist for a reachable target, and the second case is when there is no
solution and the target is unreachable. In this paper, we will show how the human
arm could simplify the IKP.
2 Methodology
Screw motion properties will be applied to the redundant robot mechanism resembles
human arm shown in Fig. 1.
Since the end-effector joint can compensate for the orientation, the mechanism
can be simplified to a two-link planar open-chain mechanism with the ability to
control the gripper’s position in space by adjusting the first two joint variables (θ1
and θ2 ). It should be noted that two cases for zero configurations of the mechanism
are to be considered. This is because it affects the optimal set of joint parameters that
minimize energy elements of the actuators. The two cases are as follows:
• Choosing the zero position to be aligned with x-axis (X0 ) (i.e., θ1 = θ2 = 0) and
• Choosing the zero position to be aligned with y-axis (Y0 ) (i.e., θ1 = 90°, θ2 = 0).
Thus, the objective is to minimize the range of angle θ1 based on the optimal
energy trajectory that the human arm can take to reach the target. Screw equation for
the simplified mechanism is given by the product of exponentials:
Bioinspired Approach to Inverse Kinematic Problem 3
Equation (1) is the homogenous transformation of the end-effector frame {n} with
respect to space frame {0}. Its derivation is based on Chasles–Mozzi theorem. In
addition, M is the transformation describing the gripper frame when the system is
at home position (i.e., case 1 or case 2). It is worth mentioning that screw motion
is grounded on a rotation (θ) about some arbitrary vector in space which is denoted
by screw axis (w) followed by a translation parallel to that axis with a distance (d).
Thus, the screw of joint is denoted by
w
S= ∈ R6 (2)
v
e[ω]θ I θ + (1 − cos θ )[ω] + (θ − sin θ )[ω]2 v
e[S]θ =
0 1
where v is the velocity vector computed as a function of a random point (q) chosen
on the axis of rotation in (3):
d
v = −[w]q + w (3)
θ
As the robot arm considered is a redundant mechanism, then dealing with singularities
is important. Singularities would define the configuration at which a motion is un-
4 R. Alkhatib et al.
executable or target that cannot be reached. Keeping in mind, some joint torques
could be bonded and in turn would result in unbounded torques by the gripper.
In Fig. 1, one can derive the position of the end effector as follows:
Taking the derivative of both (4) and (5) would lead to (6) in a matrix form as
follows:
w = J · θ −1 · Ẋ (6)
This equation can hold as long as no singularities exist. This can occur for any
configuration where changing the joints does not change the end-effector velocity in
any arbitrary direction. Mathematically speaking, this occurs when the determinant
of Jacobian (J) is zero. Otherwise, if the Jacobian is in full rank, thus the determinate
is different from zero, then the gripper can execute any arbitrary velocity.
Starting from (1), the objective is to get the joint parameters in terms of the gripper’s
configuration and is derived as in (7):
[S1 ]θ1 + [S2 ]θ2 = Log T0n · M −1 (7)
Now applying (7) for both cases of home position of the robot arms will yield two
or more scenarios that can exist for one target position as in Fig. 2. If we analyze it
graphically, it is logical to observe that the lower solution is easy to reach from the
first zero configuration as it is closer and does not require much speed, yet it still
requires some power (on actuators) to lift the mechanism. The upper solution is easy
to reach from the second configuration because it is closer, yet we notice that weight
is in the direction of motion and thus actuators consume less power despite the first
scenario.
Even though there are two solutions for one target, however, each has a different
orientation of the hand as shown in the figure above (i.e., T0n has different rotation
matrices R but the same position vector P). Since less power is the goal, obtaining
a set of angles that minimizes energy through the study of the human arm modeling
and dynamics will be considered later on. Thus rearranging (7) will be as shown in
(8):
[S2 ]θ2 = Log T0n · M −1 − [S1 ]θ1 (8)
Bioinspired Approach to Inverse Kinematic Problem 5
For example, lets us take L1 = 29 cm and L2 = 23 cm. Then the results are as
follows:
⎡ ⎤ ⎡ ⎤
0 −1 0 0 0 −1 0 29
⎢0 0 0 −29 ⎥ ⎢1 0 0 0 ⎥
S2case1 =⎢
⎣0
⎥ S2 =⎢ ⎥ (9)
0 0 0 ⎦ case2 ⎣0 0 0 0 ⎦
0 0 0 0 0 0 0 0
Then solving the IKP for θ1 ranging from –π to π and solving for θ2 will give
Table 2 in both cases as shown in Table 1, Figs. 4, and 5.
6 R. Alkhatib et al.
2.4 Analysis
Drawing the mechanisms as in Figs. 3 and 4, it clearly appears that the position of
the target is changed at each incrimination of θ1 with respect to the given orientation
α = 30° from the two cases of home position. This means that achieving a specific
orientation requires only one set of target’s position coordinates. In addition, mul-
tiple configurations with discrete incrimination of θ1 are shown in the figures. For
instance, the orientation will result in several numbers of solutions given that the
target coordinates are also specified. For illustration, Fig. 5 shows different postures.
However, choosing another target coordinates to be (0, 20, 0) will yield an infinite
number of solutions as in Table 2.
The solutions are obtained using (7). One example is shown below:
φ = π − θ2 ⇒ θ2 = π − φ
P x 2 + P y 2 = L12 + L22 − 2L1 · L2 · cos φ
⇒ φ = cos−1 (L12 + L22 − P x 2 − P y 2 /2 · L1 · L2)
α = θ1 + θ2 (Orientation)
Bioinspired Approach to Inverse Kinematic Problem 7
sin φ/(P x 2 + P y 2) = (sin β/L2) = (sin γ /L1) ⇒ β = sin−1 (L2 · sin φ/(P x 2 + P y 2 ))
The human body in general and human arm in particular is a musculoskeletal system.
It mainly consists of skeleton, skeletal muscles, and joints [4]. Knowing that, skeletal
muscle is attached to the bones by the use of tendon.
Bioinspired Approach to Inverse Kinematic Problem 9
3 Discussion
From any position of the manipulator, we can see that that joint 1 (actuator 1) carries
more load than joint 2 (actuator 2) since it is responsible for lifting and carrying
torque from the weight of two bars (Mass = 3.1 kg), despite the second actuator
which is only responsible for lifting and carrying torque of one bar (Mass = 1.1 kg).
Through observation and analysis, one would be able to say that the first joint is
dominant when we are concerned in the energy that the system is consuming as in [7,
8]. Moreover, a trajectory ranging from right to left (i.e., θ1 is increasing) consumes
the mechanism more energy than of a trajectory ranging from left to right (i.e., θ1
is decreasing). When θ1 reaches 90° and plus, the weight will start acting with the
motion of the actuator and system start requiring less energy to proceed to its final
position. Adding to this, the major factor that encloses and approves the reduced
energy configuration is time. The shortness of the path that the manipulator takes
from a certain zero configuration judges the ability of the mechanism to take the
optimum solution or not. Based on this, the following recommendations are gained:
• Target 1: The optimum solution is at θ1 = 14.8° if the trajectory starts from the
first home position, while the optimum solution is at θ1 = 95.02° if it starts from
second home position.
• Target 2: The optimum solution is at θ1 = 30.3° from first home position, while
the optimum solution is at θ1 = 83.5° from second home position.
• Target 3: The optimum solution is at θ1 = 77.8° from first home position, while
the optimum solution is at θ1 = 93.34° from second home position.
• Target 4: The optimum solution is at θ1 = 9.07° from first home position, while
the optimum solution is at θ1 = 90° from second home position.
4 Conclusion
References
1. Craig JJ. Introduction to robotics mechanics and control, 3rd edn. Pearson Prentice Hall, Upper
Saddle River
2. Hutchinson S, Vidyasagar M, Spong MW Robot modeling and control. John Wiley & Sons, Inc
3. Park FC, Lynch K (2012) Introduction to robotics
4. Winter DA (2009) Biomechanical and motor control of human movement. Wiley, New Jersey
5. Perry JC, Manning N, Burns, Blake S, Hannaford B, Rosen J (2005) The human arm kinematics
and dynamics. In: The 12th international conference on advanced robotics, Seattle WA, July
2005
6. Hansen MR, Rasmussen J, Zhou L, Bai 1 S (2011) Modeling of human arm energy expenditure
for predicting energy optimal trajectories. Denmark
7. Zhu Y, Wright W, Manocha D Conformational analysis of molecular chains using nano-
kinematics 89. North Carolina
8. Zhu Y, Wright W, Manocha D Conformational analysis of molecular chains using nano-
kinematics 90. North Carolina
Assessment of Two Musculoskeletal
Models in Children with Crouch Gait
1 Introduction
2 Method
The data used in this study was taken from the Gait and Movement Laboratory at
FLENI Institute for Neurological Research (Escobar, Argentina). These data were
collected from ten participants with CP that walked on level ground without walk-
ing aids and self-selected walking speed (Table 1). Every participant presents crouch
gait (≥25◦ of knee flexion in mid-stance), has diagnosis of symmetric Spastic Diple-
gia, has neither botulinum toxin treatment nor orthopaedic surgery within the last 6
months and has Gross Motor Function Classification System (GMFCS) levels I, II
or III. Gait Deviation Index (GDI) was used for the overall evaluation of walking.
The institution’s Research Ethics Committee approved this study. The protocol was
explained to each subject, and the participants or their caregivers provided written
informed consent.
The motion capture system included an 8-camera Elite 2002 BTS System (BTS
Bioengineering, Italy), two Kistler 9281E instrumented force plates (Kistler Group,
Switzerland) and an electromyography device Teleemg (BTS Bioengineering, Italy).
Twenty-two retro-reflective skin markers were placed following the protocol pro-
posed by Davis et al. [2] and the SENIAM recommendations [17] were followed for
the acquisition of surface electromyography signals for the rectus femoris, medial
hamstrings, anterior tibialis and gastrocnemius. Force data from the plates as well
Table 1 Description of participants and general functional characteristics of gait patterns
Subj. Gender Age Mass Leg length Velocity (m/s) Cadence Stance phase GMFCS GDI
(years) (Kg) (m) (non-dimensional)a (steps/min) (%) (Right/Left)
1 F 11 40 0.74 0.57 (0.212) 103 64 III 62.12/67.61
2 F 23 52 0.77 0.89 (0.324) 110 64 II 75.39/71.46
3 F 12 36 0.81 0.69 (0.245) 106 65 II 67.45/71.55
4 F 14 42 0.80 1.17 (0.418) 130 56 I 71.99/78.79
5 M 13 46 0.83 1.39 (0.487) 124 60 II 72.85/68.10
6 M 8 28 0.64 0.64 (0.256) 107 59 II 64.21/69.04
7 M 8 25 0.62 1.15 (0.467) 136 58 II 71.27/71.54
8 M 14 34 0.81 1.17 (0.415) 134 60 II 68.80/68.98
9 M 12 41 0.77 0.99 (0.360) 121 62 II 60.45/62.78
10 M 19 79 0.99 1.06 (0.340) 110 60 I 66.39/72.51
Mean – 13.4 42.3 0.778 0.972 (0.352) 118.1 60.8 - 68.09/70.23
Assessment of Two Musculoskeletal Models in Children with Crouch Gait
A comparative analysis between the two musculoskeletal models was performed. The
main modelling features analysed for each model are described below. In particular,
the Hill-type muscle-tendon units model proposed by Zajac [20] is used by both
musculoskeletal models.
Musculoskeletal Model 1 The first musculoskeletal model, proposed by Delp et
al. [5], is available in OpenSim and consists of 10 segments, 7 joints, 19 degrees
of freedom (Dofs) and 92 musculotendon actuators to represent 76 muscles in the
lower extremities and torso [4].
Inverse kinematics and inverse dynamics, from the experimental marker trajec-
tories and ground reaction forces were used to calculate joint angles and moments.
The residual reduction algorithm (RRA) was used to minimize the non-physiological
forces and moments applied at the pelvis by making small changes to measurements
that may contain an experimental error.
The computed muscle control algorithm (CMC) [18] was used to estimate the
muscle forces. The cost function (1), that represent the central nervous system, was
the sum of squared muscle activations:
76
J1 = (am (t))2 , (1)
m=1
1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside the
United States, check the laws of your country in addition to the terms
of this agreement before downloading, copying, displaying,
performing, distributing or creating derivative works based on this
work or any other Project Gutenberg™ work. The Foundation makes
no representations concerning the copyright status of any work in
any country other than the United States.
• You pay a royalty fee of 20% of the gross profits you derive from
the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty
payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information
about donations to the Project Gutenberg Literary Archive
Foundation.”
• You comply with all other terms of this agreement for free
distribution of Project Gutenberg™ works.
1.F.
1.F.4. Except for the limited right of replacement or refund set forth in
paragraph 1.F.3, this work is provided to you ‘AS-IS’, WITH NO
OTHER WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR ANY PURPOSE.
Please check the Project Gutenberg web pages for current donation
methods and addresses. Donations are accepted in a number of
other ways including checks, online payments and credit card
donations. To donate, please visit: www.gutenberg.org/donate.
Most people start at our website which has the main PG search
facility: www.gutenberg.org.