KEC Conference

__________________________________________________________________________________________

THE BIONIC ARM

Dolendra Sharma Abhirath Bahadur Shah Nishanta Bhattarai
Department of Computer and Electronics Department of Computer and Electronics Department of Computer and Electronics
Engineering Engineering Engineering
Kantipur Engineering College Kantipur Engineering College Kantipur Engineering College
Lalitpur, Nepal Lalitpur, Nepal Lalitpur, Nepal
[email protected] [email protected] [email protected]

Subin Thapa Surendra Shrestha, PhD

Department of Computer and Electronics Associate Professor
Engineering DOECE
Kantipur Engineering College IOE, Pulchowk Campus
Lalitpur, Nepal [email protected]
[email protected]

Abstract— The bionic arm is an electro-mechanical device II. OBJECTIVE

that can be primarily used and applied in the field of
disability. The whole device is the combination of  To design a microcontroller based bionic arm using
electrodes governed by the EMG muscle sensor that is electromyography signals derived from the skeletal
merged in a 3D printed mechanical design along with other muscles and implement it by generating various useful
smaller units. The device detects the type of muscular motions.
movement based on the voltage level provided by the EMG
sensor and according motion is performed. with the help of III. LITERATURE REVIEW
servomotors. There is a variation in the values of the voltage Until recently, robots were mainly used in factories for
level for different muscular movements. The tension that the automating production processes. In 1970s, the appearance
muscle generates during their contraction and relaxation
also creates a variation of the voltage level. Based on the
different values the thresholds were sampled and the bionic Of factory robots led to much debate on their influence on
arm based on these varying voltage level shows according employment. Mass unemployment was feared. Although
motion. this did not come to pass; robots have radically changed the
way work is done in countless factories. New robotics no
Keywords— Electromyography, EMG Muscle Sensor, longer concerns only factory applications, but also the use of
Bionic arm. robotics in a more complex manner. Several arms such as
the Bebionic 3 and iLimb are myoelectric controlled robotic
arms commercially available to the public. Numerous more
I. INTRODUCTION
prosthetic arms exist in research labs around the world,
The most valuable possession to any human being is their which are usually developed as prototypes to test advanced
body. Absence of human hand is a challenging situation, designs and concepts. Research prosthetics are generally
which makes one, truly adore the complexity of human more complex in terms of mechanical design, control, and
body. With the aim of implementing the knowledge of monitoring systems but are inferior to commercial devices
electronics field, which would primarily play a role in in terms of practicality, cost and robustness [1]. The human
improving the lives of this particular disabled group of hand comprises of 27 bones (depending on the individual),
people the project, is motivated. more than 30 individual muscles and over 100 named
Until recent times the design of such arms has progressed ligaments, nerves and arteries. Prostheses aim to replicate
relatively slowly. Early innovations such as wooden arms the functions of the human body and return functionality to
are traditional version of such arms. History shows that for a Persons with missing extremities. No current prosthetics can
long time, bionic arms have remained passive device that match the dexterity, flexibility and fluidity of the human
offer little in terms of control and movement. However, hand. We can move forward/backward, up/down, left, and
recent times have given advancements in such devices. right. At the same point, we can also rotate around 3
Slowly we are approaching an advanced trans-human different axes. The human neck for example has 3 degrees
integration between the machine and human body. of rotational freedom we can look left/right, up/down and
tilt our head sideways. Therefore, in total a single point can
have a maximum of 6 degrees of freedom (3 Translational, 3
rotational) the human finger in total has 4 degrees of
freedom. Three of these are the rotations of each joint (DIP,
PIP, MCP) which combine to control flexion and extension
of the finger. The Knuckle (MCP joint) also allows for
abduction/adduction (wiggling the finger from side to
side).In the thumb the lower CMC joint also allows for
abduction/adduction, which gives 5 DOFs in the thumb.

KECConference2019, Kantipur Engineering College, Dhapakhel Lalitpur 148

 KEC Conference
__________________________________________________________________________________________

Fingers and all joints in the human body are actuated developing 3D printed bionic limbs have become quite
(moved) via contraction of muscles and tendons [2]. popular. InMoov is an independently run project developing
a life like humanoid robot from 3D printing technology. The
The vast majority of commercial prosthetic fingers are entire project is open source and provides great mechanical
actuated through a joint linkage system powered by DC design insight into producing 3D printed robotic body parts.
electric motors. All joints in the finger are controlled We also used InMoov for the design of our bionic arm. The
through a single actuator, which means the entire finger has open source nature of this project allows the public to access
only a single degree of freedom these fingers can only computer aided designs and follow step by step guides on
open/close in a single way. In reality, a human finger has how to 3D print and assemble this system. Tendons actuated
control over individual joints so is capable of flexing in a through servomotors placed in the forearm control InMoov
variety of ways. Dexterity arises from the numerous degrees fingers. The problem with the InMoov hand is that the
of freedom of the human hand. The fine motor control a Servos take up the entire forearm leaving no room for it to
person has over their individual finger joints allows for a be attached to a stump between the elbow and wrist.
vast array of intricate tasks achieved. In contrast, The bionic arm we created focuses on those people who
commercial prostheses are limited to simple tasks partially cannot buy the expensive arms out there in the market. It
due to the lack of fine control in the fingers. For example, only performs picking and holding small objects to some
trying to knit, sew or play a musical instrument, as a guitar extent. Although there is huge level of improvements
with a modern commercial prosthetic device would be between the past designs and recent trends in the field of
extremely difficult if not impossible. Another critical design prosthetic arms, the proposed system has its own
point in commercial prosthesis is durability. The average constraints. The system depends on the muscular simulation
user will wear a myoelectric prosthetic hand in excess of 8 rather than neurological simulation in the case of recent
hours per day. Therefore, prosthetic arms for commercial advancements. Similarly, the thermal phenomena for
use must be robust, lightweight and packaged into a closed movement of the arm are not significant in the case of
system that can be attached to an amputee. Mechanical proposed system. However, the system has potential of
complexity determines the degrees of freedom in the rotational and transitional motions and the force factor for
system; however, there is usually a trade-off because holding the objects up to a certain extent.
increasing complexity can lead to an increase of the size of
the device and reduce robustness and durability [4]. The IV. PROBLEM STATEMENT
Bebionic 3 is a world leading commercial myoelectric arm. In a scenario where a person can expose ones potential to
Like others of its kind, the Bebionic 3 uses a predefined grip the fullest by learning, sharing, exploring and trying
system. A user can select from 14 different grip patterns different things, disability can be proven a curse. The
using muscle activity around their upper forearm [5]. The handicapped and disabled people are unfortunately not
user does not essentially have control of individual finger capable of performing day-to-day tasks that normal people
movements, rather they can select a grip pattern and then can perform and this consequently leads to disabled people
use muscle activity to activate the movements of that lagging behind in this modern era where there is always a
specific grip. One of the most advanced modern prosthetic struggle for existence. In this area of disability regarding
arms is the 22 degree of freedom Intrinsic Hand developed absence of arms, the proposed arm helps people to perform
at the John Hopkins Applied Physics Laboratory. This hand day-to-day actions like normal hands up to certain extent.
has been developed through DARPA initiative and funding The bionic arm we created also performs actions like
and has unmatched mechanical dexterity. To achieve such holding picking objects of small mass and dimensions to
fine control designers incorporated 15 miniature DC motors some extent such as bottles. The motion includes complete
directly in the fingers, palm and wrist. The Intrinsic Hand is opening of the hand, closing of hand in the form of fist,
able to replicate almost every movement of the biological rotating the wrist in the form of fist and pointing of index
human hand. Using standard EMG sensing techniques there finger. These motions have their own significances and
is no way of obtaining enough control for a user to purposes that can be used in day-to-day life. Moreover,
practically use all the degrees of freedom of this device. although there is availability of modern advanced prosthetic
and bionic arms, these are out of reach of majority of the
However, DARPA is further funding the development of a people and this proposed bionic arm can be implemented in
prosthesis/brain neural interface to connect the users’ an optimum cost margin thus making this technology more
nervous system directly to inputs in the arm. Another available to the people who need it. Thus, the proposed
interesting method of actuating prosthetic hands is by using bionic arm is motivated by the problems stated and is
shape memory alloys (SMAs). SMAs return to a predefined expected to solve such problems relatively along with its
shape or size when subject to the appropriate thermal own drawbacks as well.
procedure (heating or cooling).There are two major
problems with this method though designing a precise V. WORKING PRINCIPLE
control system for SMA actuators is very complex.
Contraction rate of the SMA, spring response and varying A. Materials
weights of objects grasped all contribute to this problem.  Arduino Uno
Heating the SMA and then waiting for it to cool can take  Servo Motors
quite some, which makes closing and opening a finger a  Electrodes
relatively slow process. A great benefit it that they are noise  EMG muscle sensors
free and low weight. Over the past couple of years,  The prosthetic arm

KECConference2019, Kantipur Engineering College, Dhapakhel Lalitpur 149

 KEC Conference
__________________________________________________________________________________________

B. Methodology fetched from the sensor are very low, appropriate filtering
and amplification is used to map the signals. This
amplification and filtering process is a mechanism built
within the sensor itself. The type of motion is based on these
input values and the servomotors are operated in such a way
that corresponding rotational or transitional motions are
performed. As shown in the figure the differentiating
electrodes are kept in the forearm area whereas the reference
electrodes are placed in the elbow region. The difference
read by the differentiating electrodes serves as input to the
microcontroller.
Regarding the protocols and guidelines while interfacing the
EMG sensor v3 with the microcontroller, the sensor needs to
be supplied ±18 V and a ground supply. If the supply
exceeds the EMG sensor is prone to destruction. Likewise,
the operation needs to be based on an area with low electric
field due to electrical interference. The data provided by the
EMG sensor is also dependent on the psychological state of
Fig 1: Block diagram of proposed system the host. Likewise common ground supply needs to be
arranged for reduction of unwanted oscillations.
The basic concept is to use the input derived from
electromyography data to control a robotic arm. The input
values from the muscle sensor extracted in accordance to the VI. FLOWCHART OF THE SYSTEM
amount of muscle contraction as well as the number of
contracted muscles or in other words, the stronger the
muscle contraction the higher the number of activated
muscles, the higher the recorded voltage amplitude will be.
This happens due to the contraction of muscles fibers when
the potentials are geneSrated in motor neurons. When the
neuron and axon exceed the threshold in Postsynaptic
Membrane of Neuromuscular junction, it becomes a muscle
action potential. Difference in the potentials, the muscles
potential is propagated in both directions of muscles fiber
triggering the process of the sliding of action filaments on
myosin., Various number of fibers are activated on
contractions of Muscles Unit (MU) with the combinations of
activations and synchronizations. The frequency of MU
refers to contractions, relaxing after each activation that
produces temporal pules of two or more MU firing in
combinations. Activations and synchronizations control
contractions of Muscles Unit (MU). The wrist in the
proposed arm has six degrees of freedom, the thumb having
5 degrees of freedom and the remaining 4 fingers having 4
degrees of motion individually.

Fig 2: Placement of electrodes

Fig 3: Flowchart of the system
The values read by the EMG sensor are the main working
basis for the motion of bionic arm as different values lead to
Since EMG activity is even measurable when we do not different type of motions. Initially the arm is in reset
move our body i.e. during stationary positions, a system is position, the reset position signifies the hand open condition
designed to remove such beheaviours.The input values which is taken as the reference position. According to

KECConference2019, Kantipur Engineering College, Dhapakhel Lalitpur 150

 KEC Conference
__________________________________________________________________________________________

individual basis, the readings vary and the arm needs to be

calibrated accordingly. The readings mentioned in the
flowchart are according to the muscular motion of one of
our project members. Motor one governs the wrist motion;
motor two governs the index fingers motion, motor three
controlling the thumb and motor four regulating the
remaining finger motion. If the EMG value is greater than
300 microvolts all the motors are rotated by 10 degrees that
leads to the opening of hand accompanied by the fishing
lines creating an interface between the motors and the
fingers. Similarly if the value falls on else condition i.e.
value being in between 300 micro volts and 380 micro volts,
closing of hand occurs due to rotation of motor 1 which
controls the wrist by 10 degree and other motors controlling
the fingers by 170 degrees leading to the complete closure
of palm. Likewise, if the value falls to be in between 380
microvolts and 450 microvolts all motors rotate by 170
degrees which makes the hand close and the wrist rotate
resembling of fist. Again, if the value comes to be greater
than 450 micro volts, motor 2 rotates by 10 degrees and
other motors rotate by 170 degree which finally leads to a
fist with pointed finger. The value is read constantly and
corresponding motions are performed based on the nature of
EMG values as mentioned above. The response time
between arrival of EMG signal and movement of the arm is
generally dependent on the delay parameter mentioned in
Fig 4: Graph for hand open and hand closed position
the code and processing capacity of the sensor. The
proposed bionic arm is found to have a delay of 500ms
when tested by a microsecond timer.

VII. RESULT AND DISCUSSION

The graphical representation along with the value of
readings in tabular form for various type of motion
represented by the arm is attached along. The nature of
readings given by the EMG sensors for different positions
are compared graphically. The reading observed when the
hand is at open position is taken as the reference position
and is compared to various motion of hands. The first graph
compares the nature of readings between the hand at open
position to hand at close condition whereas the second graph
compares the nature of values between the reference
position and hand being closed and rotated position.
Likewise, the third graph represents the comparison between
the reference position and the pointing index finger position.
The values can be concluded to be stable up to applicable
state. The minor fluctuations are due to practical factors
Fig 5: Graph for hand open and hand close and wrist rotate position
such as sensitivity of equipment, readings from transient
states and so on.

KECConference2019, Kantipur Engineering College, Dhapakhel Lalitpur 151

 KEC Conference
__________________________________________________________________________________________

Fig 6: Graph for hand open and index finger pointing position Fig 7: The Bionic Arm

VIII. CONCLUSION
Considering the advantage of lightweight system over a Thus, a 3D printed Bionic Arm is designed whose operation
bulky one, the arm is 3D printed keeping it ahead of the is governed by the intensity of contraction and rarefaction of
traditional arms, which have a heavy delay in range of the muscles in the hand. The operating tools for the motion
motions due to their weight factor. EMG sensor is the main are servomotors whose nature of rotation depends on the
working tool; the rotational motion of the servo motors desired movement. Thus, system is capable in the field of
placed inside the bionic arm governs the motion. The fishing Disability, hospitality industry like Restaurants and areas
line interfaces the motors and the fingers. In this way, the where normal human hands are prone to risk. Therefore, the
motions with respective EMG values govern different bionic arm can satisfy each application along with its
motions by the rotation of servomotors where the fishing limitations as well.
line serves as a link.
IX. FUTURE ENHANCEMENT
The limitations of the project serve as a scope for future
enhancement. The grip factor for capturing of objects with a
higher weight considering the force and tensile factors can be
improved. Likewise, the motion of fingers on individual
basis can be controlled in the future. The natural motion of
the wrist that is 360 degrees is limited to 180 degree in the
current pretext but it can be expanded up to 360 degrees in
the future.

KECConference2019, Kantipur Engineering College, Dhapakhel Lalitpur 152

 KEC Conference
__________________________________________________________________________________________

X. REFERENCES
[1] ] Roberto Merletti and P Di Torino. Standards for reporting emg data.
J Electromyogr Kinesiol, 9(1):3–4, 1999.
[2] AD Wibawa, N Verdonschot, JGM Burgerhof, IKE Purnama, MS
Andersen, JPK Halbertsma, RL Diercks, and GJ Verkerke. A
validation study on muscle activity prediction of a lower limb
musculoskeletal model using emg during normal walking. In
Instrumentation, Communications, Information Technology, and
Biomedical Engineering (ICICI-BME), 2013 3rd International
Conference on, pages 260–264. IEEE, 2013.
[3] Mamun Bin Ibne Reaz, MS Hussain, and Faisal Mohd-Yasin.
Techniques of emg signal analysis: detection, processing,
classification and applications. Biological procedures online, 8(1):11,
2006.
[4] Ismail Ben Abdallah, Yassine Bouteraa, and Chokri Rekik. Design
and development of 3d printed myoelectric robotic exoskeleton for
hand rehabilitation. International Journal on Smart Sensing &
Intelligent Systems, 10(2), 2017.
[5] ] Mohammad Yahya Almuhanna. Cost-effective prosthetic hand
controlled by emg. 2016.

KECConference2019, Kantipur Engineering College, Dhapakhel Lalitpur 153