1

Department of Mechanical Enigneering

Purdue school of Engineering and Technology

Design and Analysis of Bionic Arm

ME 597 / ECE 569 Introduction to Robotics System

Course Project Report

By

Archana Eadara, Ravi Teja Nalam, Samuel Attoye,

Sai Krishna Prabhala

12/15/2014

2

Abstract

The term ‘Bionics’ implies the combination of the fields Biology and Electronics.

Bionic technology is revolutionary in nature connecting the realms of man and machine

together as a single entity. It is known fact that human body is very complicated and one of its

kind. It can be repaired up to a certain extent but it cannot be remade, neither are the parts of

human body like limbs, eyes, ears, etc. Bionic technology gives hope to people who lost their

natural body parts to disease, war, accidents and other calamities by allowing them touch,

move, walk, see and hear again.

The motivation for the advent of Bionics came from this situation. The field of bionics

group the experts from biology, surgery, electronics, robotics and many more to create artificial

body parts for the human who lost their natural parts by various reasons. These artificial body

parts are naturally controllable with the help of targeted muscle reinnervation, which enables

an amputee to control motorized prosthetic devices and to regain sensory feedback.

In this project, the design and analysis of a simple 3 degree of freedom bionic arm is

presented. The movement of the Bionic/ robotic arm is analyzed with the help of parameters

from conventional robotic manipulator concepts like forward kinematic analysis, Jacobian

Matrix representation etc. An experiment is performed by prototyping a robotic arm and

animating it with the help of microcontroller. The joints of the arm are actuated using servo

motors. The results of the project are design parameters for a bionic arm, a possible arm

movement simulation.

3

Table of Contents

1. Introduction ............................................................................................................................ 5

1.1 Definitions of Some Relevant terms ................................................................................ 5

2. A Review of the Literature .................................................................................................... 6

2.1 Study of Few Previous Bionic Arms ............................................................................... 9

2.1.1 Six DOF humanoid robot arm using kinematic analysis ........................................ 9

2.1.2 A real time EMG-based Assistive Computer Interface ........................................ 10

2.1.3 The I-LIMB ........................................................................................................... 12

2.1.4 Bionic Handling Assistant .................................................................................... 12

2.2 Sensors ........................................................................................................................... 12

2.2.1 Electrical activity in muscles ................................................................................ 12

2.2.2 What it can tell us? ................................................................................................ 13

2.2.3 How to measure electrical activity in muscles? .................................................... 13

2.2.4 Types of EMG sensors .......................................................................................... 15

2.2.4.1 Surface bipolar EMG electrodes ............................................................... 15

2.2.4.2 Intramuscular EMG sensor ................................................................. 15

2.2.4.3 Surface EMG vs Intramuscular EMG ................................................. 16

2.2.5 Measurements ....................................................................................................... 16

2.2.6 EMG signal processing ......................................................................................... 17

2.2.7 Limitations ............................................................................................................ 17

2.2.8 Electrical characteristics ....................................................................................... 18

2.3 Motors ...................................................................................................................... 18

2.4 Material Selection .................................................................................................... 20

2.5 Control….................................................................................................................22

3. Challenges in Bionic Arm Research ............................................................................... 23

4

4. Objective of the Research ............................................................................................... 24

5. Design and Mathematical Modelling ............................................................................. 25

5.1 Forward Kinematic Analysis of the bionic arm ...................................................... 25

5.1.1 Ranges of variables ............................................................................................ 28

5.1.1.1 The Shoulder Joint .................................................................................... 28

5.1.1.2 The Elbow Joint ........................................................................................ 29

5.1.1.3 The Wrist Joint .......................................................................................... 29

5.2 Velocity Kinematics-Jacobian Matrix Development ............................................. 29

5.3 Computer Aided Design – Solid Modelling ........................................................... 30

5.4 Simulation ............................................................................................................... 31

6. Experimental data and setup ......................................................................................... 32

6.1 Experiment………………………………………………………………….……32

6.2 Block diagram of the experiment………………………………………..……….32

6.3 Procedure and circuit…………………………………………………………..…33

6.4 Program Code………………………………………………………………….…34

7. Results & Discussion……………………………………………………………..……….36

7.1 Mathematical Results………………………………………………………….…36

7.2 Design Results………………………………………………………………...….37

7.3 Experimental Results…………………………………………………………..…37

7.4 Explanation of the Results……………………………………………………..…38

7.5 Advantages of bionic arm…………………………………………………...……38

7.6 Limitations…………………………………………………………………..……39

7.7 Future scope…………………………………………………………………...…39

8. Conclusion……………………………………………………………………………..…..40

References

5

1. Introduction

According to Clement, Bugler and Oliver, 2011, loss of a hand can be devastating. The

primary causes of hand loss include; trauma, dysvascularity and neoplasia. Sixty-seven percent

of upper limb amputees are male. Upper limb amputations most commonly occur during the

productive working years with sixty percent between the ages of 16 and 54.

1.1 Definitions of Some Relevant terms

Bionic Arm

A next generation prosthetic device which operates more naturally than traditional

prostheses(WORLDINFO, 2014).

Bionic

Electro-mechanical device that replaces parts of the body and possess natural human

capabilities(WORLDINFO, 2014). According to NATURE, 2014, Bionics also refers to

prostheses interconnected with the nervous system and operated by myoelectric control.

Prostheses

This refers to an artificial device used to replace a missing body part("Nature," 2014). Its

functions to restore an amount of normal functioning to the amputee("Made How," 2014). The

five generic types of prostheses include;

postoperative,

initial, preparatory,

definitive and

special-purpose(Michael J. Quigley, 1988).

Myoelectric Control

This refers to the control of an artificial device by the detection of electrode onto a living bone

on electric signals initiated by the contraction of specific target muscles("Nature," 2014).

6

Osteointegration

This refers to the process of surgically grating an artificial limb onto living bone. The

process enables efficient energy transfer and fit between body and prosthesis(Communication,

2014).

2. A Review of the Literature

Similar to industrial manipulators, biological manipulators (for instance arms) are made

up of rigid links (the bones). Each joint is usually driven by several, redundant and highly

elastic and compliant actuators (muscles and tendons). Inspired by this is the design and

functionality of a bionic robot arm which consists of three joints driven by elastic and compliant

actuators. In this initial design standard springs with linear characteristics are used in

combination with electrical drives.(Klug, Von Stryk, & Mohl, 2006)

An efferent neural signal can be used to control a robotic arm and hand, and the extent

to which sensory (afferent) feedback can be provided from the robotic device to the nervous

system. Sensors placed in existing arm muscles will pick up brain signals and sends to an

amplifier and digital signal processor, which in turn sends command signals to the artificial

arm and hand. By proper designing with good accuracy and control, a brain can communicate

directly with artificial arms. Essentially, the design should allow greater control of the arm by

decoding the signals sent by the muscles so that a patient thinks to rotate his/her wrist and

artificial wrist rotates.(Shekhar, Guha, Juliet, & Kumar, 2009)

Control a machine with thoughts:

A person can successfully control multiple, complex movements of a prosthetic limb

with his or her thoughts opens up a world of possibility for amputees. The setup -- both surgical

and technological -- that makes this feat possible is almost as amazing as the results of the

procedure.

7

The "bionic arm" technology is possible primarily because of two facts of amputation.

First, the motor cortex in the brain (the area that controls voluntary muscle movements) is still

sending out control signals even if certain voluntary muscles are no longer available for control;

and second, when doctors amputate a limb, they don't remove all of the nerves that once carried

signals to that limb. So if a person's arm is gone, there are working nerve stubs that end in the

shoulder and simply have nowhere to send their information. If those nerve endings can be

redirected to a working muscle group, then when a person thinks "grab handle with hand," and

the brain sends out the corresponding signals to the nerves that should communicate with the

hand, those signals end up at the working muscle group instead of at the dead end of the

shoulder.

Rerouting those nerves is not a simple task. Dr. Todd Kuiken of the RIC developed the

procedure, which he calls "targeted muscle re-innervation." Surgeons basically dissect the

shoulder to access the nerve endings that control the movements of arm joints like the elbow,

wrist and hand. Then, without damaging the nerves, they redirect the endings to a working

muscle group. In the case of the RIC's "bionic arm," surgeons attach the nerve endings to a set

of chest muscles. It takes several months for the nerves to grow into those muscles and become

fully integrated. The end result is a redirection of control signals: The motor cortex sends out

signals for the arm and hand through nerve passageways as it always did; but instead of those

signals ending up at the shoulder, they end up at the chest.

To use those signals to control the bionic arm, the RIC setup places electrodes on the

surface of the chest muscles. Each electrode controls one of the six motors that move the

prosthetic arm's joints. When a person thinks "open hand," the brain sends the "open hand"

signal to the appropriate nerve, now located in the chest. When the nerve ending receives the

signal, the chest muscle it's connected to contracts. When the "open hand" chest muscle

contracts, the electrode on that muscle detects the activation and tells the motor controlling the

8

bionic hand to open. And since each nerve ending is integrated into a different piece of chest

muscle, a person wearing the bionic arm can move all six motors simultaneously, resulting in

a pretty natural range of motions for the prosthesis.(Layton)

The development and availability of myoelectric prosthetic limbs, has found impotent

use in rehabilitative medicine. According to ("Made How," 2014) myoelectric research began

in West Germany 1940 and by 1960 was being applied in the development of artificial limbs.

In myoelectric prostheses, residual muscles (for a prosthetic arm biceps and triceps) serve as

natural batteries, initiating and sending signals through the skin (transcutaneous). These signals

control the movements of the prosthetic arm.

Central and peripheral motor and somatosensory pathways retain significant residual

connectivity and function for many years after limb amputation and this property has been

exploited by researchers using a technique called targeted motor reinnervation to increase the

accuracy of myoelectrically controlled prostheses(Oliver, 2011).

Further recommendations by Clement, Bugler and Oliver, 2011 suggest implantation

of bipolar differential electromyographic (EMG) electrodes within the muscle to create a

system capable of reading intra muscular EMG signals that increases the number of control

sources available for prosthesis control to increase accuracy of myoelectric prostheses.

However they claim that use of intraneural electrodes presents the most viable means for

integrating bionic limbs into the biological system. Intraneural electrodes interface directly into

the nerves in the limb stump and have the ability to carry a bidirectional flow of information

between the bionic limb and patient.

According to Quigley, 1988; factors to consider in prescribing a prosthesis include;

weight bearing

suspension

activity level

9

general prosthesis structure

components expense

certain unique considerations

2.1 Study of Few Previous Bionic Arms

2.1.1 Six DOF humanoid robot arm using kinematic analysis

These type of humanoid robot bionic arms (Figure 2.1.1a) are based on servo motors (Figure

2.1.1b). They perform the given tasks following the trajectory generated from the geometric

analysis. They are based totally on length of the robot arm and rotation angle as well. The PIC

controller 18F4520 is the main controller used with the help of USB interface.(Ohol, 2014)

Figure 2.1.1a: Humanoid robot arm Figure 2.1.1b: RC servo motor

They are mostly used in services related to transportation and welding. The advantages of these

types of robotic arms are highly accurate, highly precise and repeatability. The con about this

robot is that the workspace is so limited and so they have been trying to make a movable base

just to increase the work space envelope to make better use of it. The PIC controller used in

this has about 36 I/O pins for more robust operations and with 256kb ROM memory. Having

bevel type of transmission gears of motion at every joint makes it more robust and very

compact. The parts are made up of aluminum sheet for more strength and less weight and the

thickness is about 2mm. The length of total arm from shoulder to elbow is 240mm and from

elbow to wrist, it is about 290mm.

10

The humanoid robot arm system structure is shown in figure 2.1.1c and the structure of robot

arm in figure 2.1.1d.

Figure 2.1.1c: Humanoid robot arm system Figure 2.1.1d: Structure of six DOF robot arm

The position of the robotic arm is defined by kinematics control.

Just by adding a proximity sensor and camera, environmental information can be easily

acquired and helps to avoid obstacle autonomously.

2.1.2 A real time EMG-based Assistive Computer Interface for the Upper Limb Disabled

This is a design for the upper limb disabled. A real time assistive system to access a computer

machine via residual muscle activities without the

help of standard computer interfaces like mouse and

the keyboard (Figure 2.1.2a). In this computerizing

world with all the automation on this planet, things

would be a lot easier if they can operate a computer

even when they are disabled.

Figure 2.1.2a: Conceptual diagram of the developed EMG-based computer interface

11

In this design, the electromyogram (EMG) senses very low signals produced by the

muscle in lower arm. Then they are properly filtered before they are considered as being fed

back to the system. Operates to click and control the movement of the cursor from the signals

obtained. It also has the on-screen keyboard for entering text in different languages.(Choi &

Kim, 2007)

The locations of electrodes on the skin of the lower arm to observe EMG signals in

shown in figure 2.1.2b. But the signals received from the EMG from the central nervous system

extracting human thoughts and recording

brain activities with the help of

electroencephalogram (EEG) is very

noninvasive, low spatial resolution and has

very low Nosie to signal ratio.

Figure 2.1.2b: Myoelectric sites fur extraction of EMG signals

So, this is definitely not perfect and has a lot of issues and still showing considerable progress

from previous robots of these types. For a better signal to

Noise ratio they have started using other techniques such as

signals from the peripheral nervous system through

electromyogram and can be measured more safely and

conveniently at a CNS level. It is always better that

electromyogram based human computer interaction is more

reliable and practical with current technology.

Figure 2.1.2c: The on-screen keyboard

12

The developed on-screen keyboard to help the upper limb disabled enter Roman and

Korean letters on a computer is shown in figure 2.1.2c.

The possible extension of this project could be control of various things such as

prosthetic arms, bionic robot and disabled limbs like exoskeletons.

2.1.3 The I-LIMB

Developed commercially by TOUCH BONICS, a Scotland-based firm, with the

objective of improving prostheses power, precision and aesthetics. The I-LIMB prosthetic arm

permits the user to adjust tactile control of the device, it has five individually powered digits.

The design utilizes independently controlled miniature motors connected to each arm, wrist

and finger joints. Arm control is achieved by software controlled high-frequency electronic

pulses to individual motors and an integral microprocessor which translate the electric signals

from the forearm. A stall detection system optimizes tactile and power resources(Pavic, 2010).

2.1.4 Bionic Handling Assistant

Developed for commercial purpose by FESTO, a German-based company, the Bionic

Handling Assistant performs more of industrial handling functions. It geometric structure is

flexible and trunk-like with a multi-jointed design(Pavic, 2010).

2.2 Sensors

2.2.1 Electrical activity in muscles

A muscle, made of hundreds of cells called muscle fibers, moves due to contraction of

these fibers. Our nervous system sends electrical signals via neurons (motor neurons) to the

muscle fiber to make it contract. A single motor neuron and the fibers that are attached to it are

called a motor unit- the fundamental unit of muscle motion. Some motor neurons connect to

only a few fibers while others connect to a large number. This gives us a varying degree of

control over the different muscles (more fibers attached=less control)("Electromyography

(EMG),").

13

Figure 2.2.3a: Electromyography Figure 2.2.3b: Electromyogram

We can measure the differential voltage from two electrodes on the skin and amplify the signal.

2.2.2 What it can tell us?

Occurrence of muscle contractions

Strength of muscle contractions

2.2.3 How to measure electrical activity in muscles?

Electromyography (EMG) is a method for evaluating and recording the electrical activity

produced by skeletal muscles(Robertson, Caldwell, Hamill, Kamen, & Whittlesey,

2013)(Figure 2.2.3a). EMG is performed using an instrument called an electromyograph, to

produce a record called an electromyogram(Figure 2.2.3b). An electromyograph detects the

electrical potential generated by muscle cells when these cells are electrically or neurologically

activated. The signals can be analyzed to detect medical abnormalities, activation level, or

recruitment order or to analyze the biomechanics of human or animal movement.

14

Figure 2.2.3c: Simplified block diagram of surface electromyogram acquisition.

Block diagram (Figure 2.2.3c) showing each of the main steps regarding the acquisition of

surface electro-myograms:

1. The detection of myoelectric potentials with surface electrodes and a reference

electrode, schematically illustrated on the medial epicondyle of the humerus;

2. The amplification of such potentials with differential amplifiers;

3. Analog filtering of the amplified potentials to avoid aliasing and, finally;

4. The sampling of the surface electromyogram into digital voltage values.

5. The digital values are stored on a computer.

EMG signals are used as a control signal for prosthetic devices such as prosthetic hands,

arms, and lower limbs. EMG sensors are used to measure these EMG signals which are the

nerve impulses. These impulses are then analyzed to generate the reference torque and position

values and send it to the motion control unit.(Ebrahimi, Minzenmay, Budaker, & Schneider,

2014)(Figure 2.2.3d). There are two kinds of EMG sensors in widespread use: surface bipolar

EMG and intramuscular (needle and fine-wire) EMG.

15

Figure 2.2.3d: Targeted muscle

re-innervation in person with

shoulder disarticulation.

EMG = electromyogram

N. = nerve

2.2.4 Types of EMG sensors

2.2.4.1 Surface bipolar EMG electrodes(Figure 2.2.4a), used for recording, will be placed on

the user’s skin(Cram, Kasman, & Holtz). The raw EMG recordings were preprocessed, i.e.,

full-wave-rectified, low-pass-filtered, and normalized to their maximum voluntary isometric

contraction value(Zajac, 1988).

Figure 2.2.4a: Surface EMG electrodes Figure 2.2.4b: Intramuscular EMG sensor

2.2.4.2 Intramuscular EMG sensor is a needle electrode or a needle containing two fine-wire

electrodes which is inserted through the skin into the muscle tissue (Figure 2.2.4b). A trained

professional (such as a neurologist, physiatrist, chiropractor, or physical therapist) observes the

electrical activity while inserting the electrode. The insertional activity provides valuable

information about the state of the muscle and its innervating nerve. Normal muscles at rest

make certain, normal electrical signals when the needle is inserted into them. Then the electrical

activity when the muscle is at rest is studied. Each electrode track gives only a very local picture

16

of the activity of the whole muscle. Because skeletal muscles differ in the inner structure, the

electrode has to be placed at various locations to obtain an accurate

study.("Electromyography,")

2.2.4.3 Surface EMG vs Intramuscular EMG

Surface EMG Intramuscular EMG

The ideal location of the muscle is choosen

and the surface EMG is placed on the skin

covering that location of the muscle.

A needle electrode or a needle containing two

fine-wire electrodes is inserted through the

skin into the muscle tissue.

A surface electrode may be used to monitor

the general picture of muscle activation.

Activity of only a few fibres as observed

using an intramuscular EMG.

These can only measure superficial muscles. For deep muscles, these are intrusive and

painful.

2.2.5 Measurements

A motor unit is defined as one motor neuron and all of the muscle fibers it innervates.

When a motor unit fires, the impulse (called an action potential) is carried down the motor

neuron to the muscle. The area where the nerve contacts the muscle is called the neuromuscular

junction, or the motor end plate. After the action potential is transmitted across the

neuromuscular junction, an action potential is elicited in all of the innervated muscle fibers of

that particular motor unit. The sum of all this electrical activity is known as a motor unit action

potential (MUAP). This electrophysiological activity from multiple motor units is the signal

typically evaluated during an EMG. The composition of the motor unit, the number of muscle

factors affect the shape of the motor unit potentials in the myogram.

17

2.2.6 EMG signal processing

Rectification is the translation of the raw EMG signal to a single polarity frequency (usually

positive). The purpose of rectifying a signal is to ensure the raw signal does not average zero,

due to the raw EMG signal having positive and negative components. It facilitates the signals

and process and calculates the mean, integration and the fast Fourier transform (FFT). The two

types of rectification of signals refer to what happens to the EMG wave when it is processed.

These types include:

1. Full-length rectification

2. Half-length rectification

Full length rectification adds the EMG signal below the baseline (usually negative polarity)

to the signal above the baseline making a conditioned signal that is all positive. This is the

preferred method of rectification because it conserves all signal energy for analysis, usually in

the positive polarity.

Half-length rectification deletes the EMG signal below the baseline. In doing so, the average

of the data is no longer zero therefore it can be used in statistical analyses. The only difference

between the two types of rectification is that full-wave rectification takes the absolute value of

the signal array of data points.

2.2.7 Limitations

The actual placement of the electrode can be difficult and depends on a number of

factors, such as specific muscle selection and the size of that muscle.

Although EMG is more effective on superficial muscles as it is unable to bypass the

action potentials of superficial muscles and detect deeper muscles.

Deep muscles require intramuscular wires that are intrusive and painful in order to

achieve an EMG signal.

18

Adipose tissue (fat) can affect EMG recordings. The more body fat an individual has,

the weaker the EMG signal.

Muscle cross talk occurs when the EMG signal from one muscle interferes with that of

another, limiting reliability of the signal of the muscle being tested.

2.2.8 Electrical characteristics

The electrical source is the muscle membrane potential of about –90 mV. Measured

EMG potentials range between less than 50 μV and up to 20 to 30 mV, depending on the muscle

under observation.

Typical repetition rate of muscle motor unit firing is about 7–20 Hz, depending on the

size of the muscle, previous axonal damage and other factors. Damage to motor units can be

expected at ranges between 450 and 780 mV.

2.3 Motors

Proper selection of motor is an important aspect of this prosthetic arm design because

powering all the degrees of freedom can be done by the same type of motor. Table 2.3a explains

the comparison of different motors for normal robots to robots which needed for high power

applications as well. Also, table 2.3b gives detail differences about the pros and cons of widely

used motors.

19

Table 2.3a: Comparison of different types of motors that can used for prosthesis.

Table 2.3b: Differences of widely used motors about pros and cons

The motors we have used in our project are servo. Because when we have tried to use

D.C. motor it needs complex setup for speed control and high current for high torque which is

20

pretty hard to deal with when it comes to prototype robot arm. So, for all the requirements we

had we ended up using servo motors which are really small but powerful enough to be used in

various applications ranging from toy cars to robots. A usual servo motor can be seen as a

combination of three basic parts:

1. A controller that is used to command and to send control signals,

2. an electric motor and

3. A potentiometer feedback system which is connected to the output shaft.

The servo motor is being controlled through the pulse width modulated signal and

rotates to the specific angles in correspondence to its current angular position i.e. basing on the

duty cycle of the receiving signal. Thus by controlling the PWM and duty cycle we can

precisely choose the speed and angle of the rotation which is greatly useful in prosthetics.

2.4 Material Selection

Advances in biomaterials aim to reduce the weight of bionic prosthesis; weight

reduction improves arm manipulation(Oliver, 2011). Required electrical conductivity, good

mechanical strength, lightness and chemical inertness are important properties required for

prosthetic devices. Modern plastics (polyethylene, polypropylene, acrylics, and polyurethane)

are used for the arm structure. Alloys of titanium and aluminum with required mechanical

properties and more recently, carbon-fiber, are used for the pylons. Ppolypropylene is preferred

as the socket material.

Table 2.3 present a comparison of materials used in bionic arm manufacture. Elasticity

is avoided as it causes unwanted oscillations and increases difficulty in precise movement

control(Moehl, 2000). Inherently conducting polymers (ICP), such as polypyrroles,

polythiophenes and polyanilines due to their ability the ability to electronically control a range

of physical and chemical properties(Wallace, 2007), are used as electrode materials.

21

Table 2.3: Comparison of Some Bionics Prosthesis materials

MATERIAL CHARACTERISTICS

OF INTERERST

METHOD OF

APPLICATION

APPLICATION REMARKS+

IMPROVEMEN

TS ON

CHARACTERIS

TICS

Nano

composite

Polymer

Impervious to ionic

solutions in human body,

Ability to conduct

electricity

Stem cell

technology

Cardiovascular

Implants

Carbon-fiber

composites

:Graphene

Impervious to ionic

solutions in human body,

Ability to conduct

electricity

Bioelectronics+

Nanotechnology

+ Chemical

Vapor

Deposition

Cardiovascular

Implants

Biological

sensing- analogue

applications (Ear

and eye implants)

Cannot be

switched like

Silicon thus poor

digital application

Silicon+ Metal

Oxide

Stabilizer

Ability to conduct

electricity

Bioelectronics Oxide layers trap

ions causing

interference

Diamond

nanocrystals

Retinal implants Inflexible, poor

conductivity

Advanced

plastics

Aluminum

and Silicon

Oxide

Substrates

Pt Metallization

and firing

(Electrode forge

welding in a

metallized

feedthrough

Modern

plastics?

Polypropylene,

polyethylene,

polypropylene,

acrylics, and

polyurethane

Stronger and more

lightweight

Applied in all

joints

Titanium

Alloys

Stronger and more

lightweight

Acrylics fibers Higher durability than

polyester resins high

malleability, water

resistant

Prosthetic socks

Polyester resin Laminations in

prosthetics

Icps Good conductivity,

lightweight

22

2.5 Control

Control of industrial manipulators generally is hierarchical with a trajectory planning

phase resulting in set-point trajectories for the individual joints followed by an independent

PID joint control. The control approach for stiff robots with elastic deformations in the joints

can also be assigned to the bionic robot with its elastically coupled drive. If for specific

applications the path of the manipulator is given or prescribed in advance it is possible to

calculate an optimized trajectory which is time optimal and which compensates the oscillations

within the feed-forward term (Klug, Von Stryk, & Mohl, 2006).

The neural input to the muscles and the joint angles were measured from two subjects

(2 male, age 21, 23) while they made a point-to-point movement with their right index finger

We used a PHANToMTM Premium 1.5 robot (Sensable Technologies, Inc.) to record the index

finger movement. This robot has 3 degrees of freedom and a custom-made finger cuff provided

an additional two rotational degrees of freedom. The PHANToM recorded the Cartesian

position of the robot endpoint, and we attached a potentiometer to the pitch axis of the cuff to

record the pitch of the fingertip.

The subjects’ index finger was fastened into the finger cuff so that their distal

interphalangeal joint (DIP) was aligned to the edge of the cuff. The Cartesian position of the

DIP was determined by calculating its geometric relationship with the PHANToM endpoint.

The subjects’ palm and other digits were strapped to an armrest so that all the joints on the

index finger could move freely while the hand was fixed a known distance away from the

robot’s origin. We recorded the joint angles of the robot and the finger cuff at 1 kHz. (Pedram

Afshar, Yoky Matsuoka).

23

3. Challenges in Bionic Arm Research

Achieving fine motor control (the simultaneous use of multiple joints, or full rotation)

is difficult, due to the inability to “feel” the prosthesis.

The socket interface used to attach the prosthesis may interfere with the function of a

residual joint such as the elbow.

Also there is the susceptibility of osseous-integrated device to infection at the

prosthesis-skin interface.

The prosthesis also requires extensive training and occupational therapy(Oliver, 2011).

Other challenges in bionics research are in electrode manufacturing; according to

Schmidt, 2012; chemical vapor deposition, doesn't generate perfect graphene, and this

limits the material's electronic performance.

The signals received from the EMG from the central nervous system are noninvasive,

low spatial resolution and has very low Nosie to signal ratio.

Training the subject to overcome the additional stress put on the brain by the control of

the bionic arm.

To prevent the adverse unwanted genealogical defects in the offspring the test subject.

24

4. Objective of the research

Main objective: The main objective of the project is to analyze and explain the working of a

Bionic arm and to design a prototype robotic arm to simulate its working using potentiometers

as analog sensors.

Other objectives of the project:

In order to achieve the main objective, the project was directed through a series of some

secondary objectives. They are stated below in order.

To analyze and decide number of degrees of freedom of a bionic arm.

To perform forward kinematic analysis, Jacobian matrix development which is crucial

to determine the velocity kinematics of the arm and end effector.

The sensors are the main inputs to the bionic arm. In order to simulate the sensing,

potentiometers are to be used as analog sensors instead of conventional EMG sensors

as their operation is identical.

To build a light weight, prototype 3 DOF robotic arm to simulate the working of bionic

arm using servo motors.

Selection of a microcontroller to program the servo motors of the robotic arm.

To build the circuit so that the microcontroller takes inputs from the sensors

(potentiometers) and move the servo motors of the arm based on the position of the

knob of the potentiometer.

Finally, to relate the working of this prototype robotic arm and the setup to the actual

functioning of the bionic arm.

25

5. Design and Mathematical Modelling

5.1 Forward Kinematic Analysis of the bionic arm: The Denavit-Hartenberg Convention

Problem: To obtain the relationship between the individual joints of the robotic arm and the

position and orientation of the end-effector.

For this, forward kinematic analysis has been done on the bionic arm. The joint

variables here are the angles between the links. We incorporated a commonly used convention

for selecting frames of reference the robotic arm which is the Denavit-Hartenberg (DH)

Convention (Mark W. Spong, 2006). Here each homogeneous transformation Ai is represented

as a product of four basic transformations.

Here the four parameters associated with link i and joint i. These are named as,

= link length

= link twist

= link offset

= joint angle

As all the three joints of a bionic arm are revolute, the only joint variable is joint angle

(θi). The other three parameters are constant.

26

The joint axis Z0, Z1, Z2 and X3(n) are normal to the page. We established the base

frame as O0X0Y0Z0 at the shoulder joint. The intersection point of the Z0 axis with the page is

chosen as the origin. X0 axis direction is chosen completely arbitrary. Once we established the

base frame at shoulder joint, the frame O1X1Y1Z1 at

elbow joint is fixed. Here the origin O1 of the frame is

located at the intersection of the Z1 with the page. Then

the frame O2X2Y2Z2 wrist joint is fixed. Now, the origin

O2 is chosen at the intersection of Z2 with the page.

The final frame O3X3Y3Z3 for the gripper is

fixed by choosing the origin O3 at the end of link 3 as

shown in the figure 5.1.The DH parameters are

calculated and shown in the Table 5.1.

Figure 5.1: Bionic arm.

The Z0, Z1, Z2, X3(n) all point out of the page, and are not shown in the figure.

Table 5.1 Link parameters for bionic arm

Links ai di αi θi

Link 1 a1 0 0 θ1*

Link 2 a2 0 0 θ2*

Link 3 0 0 90o θ3*

* variable

Where,

ai = distance along xi from oi to the intersection of the xi and zi-1 axes.

di = distance along zi-1 from oi-1 to the intersection of the xi and zi-1 axes. di is variable if joint i

is prismatic.

αi = the angle between zi-1 and zi measured about xi.

27

θi = the angle between xi-1 and xi measured about zi-1. θi is variable if joint i is revolute.

Now using the DH parameters, the homogeneous transformation matrices or simply A-matrices

A1, A2, A3 are determined from,

Ai =

1000

cossin0

sinsincoscoscossin

cossinsincossincos

iii

iiiiiii

iiiiiii

d

a

a

A1 =

1000

0100

sin0cossin

cos0sincos

1111

1111

a

a

A2 =

1000

0100

sin0cossin

cos0sincos

2222

2222

a

a

A3 =

1000

0010

0cos0sin

0sin0cos

33

33

The Transformation matrix or simply T matrices have been determined using A-matrices. They

are given by

T10 = A1

T20 = A1 A2

T30 = A1 A2 A3

=

1000

0010

sinsincos0sin

coscossin0cos

11122123123

11122123123

aa

aa

28

We noticed the first two entries of the last column of T30 are the x and y components

of the origin O3 of the gripper frame with respect to the base frame i.e., the shoulder frame.

That is,

X = 11122 coscos aa and

Y = 11122 sinsin aa ,

are the coordinates of the end-effector i.e., gripper in the base frame i.e., shoulder frame. Also,

the rotational part of the T30 gives the orientation of the gripper (end-effector) frame O3X3Y3Z3

relative to the base frame which is the shoulder frame.

5.1.1 Ranges of variables: Human Anatomy-Joint Limitations

The ranges of the variables in the DH table are determined from the human arm joint

limitations.

5.1.1.1 The Shoulder Joint

Of all the joints, this one has the biggest range.

It can be shown that the shoulder joint can rotate

vertically from 0o in rest position to 180o in forward

direction and to 60o in backward direction (Figure

5.1.1a). Hence the range of rotation is up to

240o.(Medlej, 2014)

Figure 5.1.1a: Shoulder joint limitation

29

5.1.1.2 The Elbow Joint

In stark contrast to the shoulder, the elbow is like a door hinge –

it opens in one direction and meets a stop (Figure 5.1.1b).

Therefore, an elbow joint can rotate up to an angle of

150o.(Medlej, 2014)

Figure 5.1.1b: Elbow joint limitation

5.1.1.3 The Wrist Joint

The wrist's range of motion is almost all front

and back; if we try to rotate it, we notice that it does not

describe a proper circle, but more of an ellipse, because it

can move so little to the sides. In our bionic arm analysis,

we considered only the front and back motion. Figure 5.1.1c: Wrist

joint limitation.

As shown in the figure, the range of rotation is 80o-90o forward and 70o backward.

Therefore, the angle of rotation varies from 0o to 150o-160o.(Medlej, 2014)

5.2 Velocity Kinematics-Jacobian Matrix Development

The Jacobian matrix relates the vector of joint velocities to the body velocity. This

matrix is an important quantity in the analysis and control of robot motion(Mark W. Spong,

2006).

The Jacobian matrix for the arm is given as;

30

𝐽𝑎𝑟𝑚 = [𝐽𝑣𝑖𝐽𝜔𝑖

] = ⌊𝑧0 × (𝑜3 − 𝑜0 ) 𝑧1 × (𝑜3 − 𝑜1 )

𝑧0 𝑧1

𝑧2 × (𝑜3 − 𝑜2)𝑧2

]

From earlier developed A matrices;

𝑧𝑜 = [001] ; 𝑧1 = [

001] ; 𝑧2 = [

001]

𝑜𝑜 = [001] ; 𝑜1 = [

𝑎1𝑐𝑜𝑠𝜃1

𝑎1𝑠𝑖𝑛𝜃1

0

] ; 𝑜2 = 𝑜3 = [𝑎2𝑐𝑜𝑠(𝜃1 + 𝜃2) + 𝑎1𝑐𝑜𝑠𝜃1 𝑎2𝑠𝑖𝑛(𝜃1 + 𝜃2) + 𝑎1𝑠𝑖𝑛𝜃1

0

]

Giving the Jacobian of the arm as;

𝐽𝑎𝑟𝑚 =

[ −𝑎2𝑠𝑖𝑛(𝜃1 + 𝜃2) − 𝑎1𝑠𝑖𝑛𝜃1 −𝑎2𝑠𝑖𝑛(𝜃1 + 𝜃2) 0𝑎2𝑐𝑜𝑠(𝜃1 + 𝜃2) + 𝑎1𝑐𝑜𝑠𝜃1 𝑎1𝑐𝑜𝑠𝜃1 0

0 0 00 0 00 0 01 1 1]

5.3 Computer Aided Design

For any mechanical/ electromechanical project, a good solid design model is very

crucial. It helps in visualizing the device or the mechanism. Several highly efficient modelling

software programs are used all over the world. Here also, we used an efficient solid modelling

software called ‘Creo Parametric’ developed by PTC. The as built solid model of the proposed

3 Degree of freedom bionic arm is shown in figure 5.3.

31

Figure 5.3.1: Creo Parametric Solid model of the Bionic arm

5.4 Simulation:

The above shown bionic arm solid model has several mechanisms at the joins, wrist

and the gripper. They are all constrained to a particular movement using the mechanism tool

in the software. All the joints are pin constrained and servo motor constraints are used to

simulate the movements of the arm. With the help of Creo Parametric, we can simulate the

mechanical joints by means of virtual actuators and it can generate a video file with a suitable

aspect ratio and frame rate. Figure 5.4 is a snapshot from the video which depicts the motion

of the bionic arm solid model.

32

Figure 5.4.1: Snapshot of the video showing the motion of the bionic arm solid model.

6. Experimental data

6.1 Experiment:

Aim of the experiment: To interpret the movement of a Bionic arm by animating a prototype

robotic arm using servo motors and microcontroller by varying resistance using potentiometers.

6.2 Block Diagram:

The block diagram of the experimental setup is shown below. It consists mainly of

Arduino Uno Atmel series microcontroller, 3 potentiometers of range 10K Ohms, 3 Servo

motors and a power supply.

33

Figure 6.2.1: Block diagram of the experimental Setup

6.3 Procedure and Circuit:

The step by step procedure to perform this experiment is given below:

The circuit for the above shown setup is shown below.

The microcontroller is powered by a computer using a standard USB port delivering 5V. It

has analog I/O pins from A0 to A5 and digital I/O pins from 0 to 13.

The potentiometers are analog sensors which are connected to the A0, A1 and A2 pins.

The microcontroller obtains the analog input from the potentiometers and convert them into

digital using an in built Analog to Digital Converter.

These digital outputs are sent to the digital pins 9, 10, 11 through a set of instructions via

programming.

The digital outputs are connected to the corresponding control signal ports of the respective

servo motors. The servo motors and the potentiometers are connected to 5v and ground

GND.

When the circuit is powered, whenever a potentiometer is operated across its resistance

range, a corresponding servo motor moves in the range of 0 to 170 degrees angle of rotation

and helps in moving the corresponding joint link of the robotic arm.

The potentiometers and the servo motors are sensitive enough and programmed with less

delay in order to obtain sharp and accurate response.

34

Figure 6.3.1: Circuit diagram of the experiment setup

The results of this experiment are the calibrated movement of the arm joints with the

help of servo motors whose angle is controlled by the potentiometer which is an analog

sensor. The microcontroller is programmed by the following program code to do these

operations.

6.4 Program Code:

// controlling servo position using a potentiometer (variable resistor)

#include <Servo.h>

Servo servo1; // create servo object to control a servo

Servo servo2;

Servo servo3;

int potpin1 = 0;

int potpin2 = 0;

35

int potpin3 = 0; // analog pin used to connect the potentiometer

int v1=0;

int v2=0;

int v3=0; // variable to read the value from the analog pin

void setup()

{

servo1.attach(9); // attaches the servo on pin 9 to the servo object

servo2.attach(10);

servo3.attach(11);

}

void loop() // for continuous iterations

{

v1 = analogRead(potpin1); //reading the analog input from pin A0

v1 = map(v1, 0, 1023, 0, 160); // mapping the ranges of potentiometer and servo motor

servo1.write(v1); // assigning the position to the servo based on input v1

delay(15); // time to respond for the servo motor for accuracy and tuning

v2 = analogRead(potpin2); //reading the analog input from pin A0

v2 = map(v2, 0, 1023, 0, 160);

servo2.write(v2);

delay(15);

v3 = analogRead(potpin3); //reading the analog input from pin A0

v3 = map(v3, 0, 1023, 0, 160);

servo1.write(v3);

delay(15);

}

36

7. Results & Discussion

7.1 Mathematical Results:

To know the position and orientation of the end effector for any given joint variables we used

the results of forward kinematic analysis.

T30 = A1 A2 A3

=

1000

0010

sinsincos0sin

coscossin0cos

11122123123

11122123123

aa

aa

For instance, consider

Upper arm length 1a = 30cms, Forearm length 2a = 30cms

Shoulder rotation 1 = 90o, Elbow rotation 2 = 30o, and wrist rotation 3 = 45o

The end-effector or gripper position of the bionic arm can be obtained as

Tgripper0 =

1000

0010

418.47066.00998.0

425.24998.00066.0

For the same sample data, the Jacobian matrix can be obtained as

𝐽𝑎𝑟𝑚 =

[ −𝑎2𝑠𝑖𝑛(𝜃1 + 𝜃2) − 𝑎1𝑠𝑖𝑛𝜃1 −𝑎2𝑠𝑖𝑛(𝜃1 + 𝜃2) 0𝑎2𝑐𝑜𝑠(𝜃1 + 𝜃2) + 𝑎1𝑐𝑜𝑠𝜃1 𝑎1𝑐𝑜𝑠𝜃1 0

0 0 00 0 00 0 01 1 1]

37

𝐽𝑎𝑟𝑚 =

[ −47.418 −17.418 0−24.425 0 0

0 0 00 0 00 0 01 1 1]

7.2 Design Results

7.3 Experimental Results:

The as built experimental setup of microcontroller, servo motors, potentiometers and other

circuit elements is shown below.

With the help of the circuit shown below, the prototype robotic arm is succesfully

simulated.

The Arduino Controller is programmed in a way that when a potentiometer is

operated, the corresponding servo motor is moved. There are totally 3

potentiometers and 3 servo motors.

38

Figure: Circuit of the experiment

7.4 Explanation of experimental results:

The experimental setup is analogous to a general 3 DOF bionic arm.

The potentiometers in this circuit operate just like EMG sensors. EMG sensors take

input from the brain through the motor nerves. Since we cannot provide brain signal

as input, we vary the resistance of the potentiometers to animate the arm with servo

motors.

Hence the motors operate based on the input of the sensors which is identical to the

operation of the bionic arm.

7.5 Advantages of Bionic arm

There are wide range of benefits using prosthetic devices, including a more natural

appearance and more control over the area of the body with missing limb or part.

The obvious benefit is to improve the quality of life for those with certain disabilities.

An amputee can have freedom of movement with more natural ease.

An amputee can use less amount of energy for the same work.

39

Versatility in use, Handle many grip functions, Looks and moves almost like a real hand.

7.6 Limitations

The major downsides to using prosthetic devices are the risk of complications.

These bionic arms are really expensive, vulnerable to repairs and quite heavy.

This artificial body parts can cause a negative genealogical effect on future generations

Many can cause skin irritation at the sight where they are fitted, and most patients undergo

some level of physical therapy as they get used to using the new device.

The stress on brain is more than usual, since it is necessary to retrain and re-learn how to

use certain muscle groups in many cases.

7.7 Future Scope

• The joint motors need more tuning and sensitivity need to be improved a lot.

• Advent in EMG sensor technology can really improve the accuracy of arm motion.

• Bionic Arm can be equipped with force sensors to acquire the ‘sense of touch’.

• Future advanced bionic arms can improve the human capacity by many times.

40

9. Conclusion

Patients requiring amputations present a significant portion in need of rehabilitative

medicine. To treat them, scientists continue to pursue research and development of bionic arms

and hands with full motor and sensory function(Surgeons, 2014). The lifelike prosthesis helps

the patient emotionally, socially and professionally. According to Clement, Bugler and Oliver,

2011; improving technology has a high probability of raising demand for and application of

bionic hands. Therefore a wider understanding is required. Advances in motors, gearboxes,

batteries and electronics will facilitate further breakthroughs in prosthetic design(Pavic, 2010).

41

References (hyperlink activated)

Choi, C., & Kim, J. (2007). A real-time EMG-based assistive computer interface for the upper limb disabled. Paper presented at the Rehabilitation Robotics, 2007. ICORR 2007. IEEE 10th International Conference on.

Communication, H. (2014). Popular Mechanics. from http://www.popularmechanics.com/ Cram, J., Kasman, G., & Holtz, J. Introduction to surface electromyography. 1998. Gaithersburg,

Marland: Aspen publishers Inc. Ebrahimi, A., Minzenmay, D., Budaker, B., & Schneider, U. (2014). Bionic upper orthotics with

integrated EMG sensory. Paper presented at the Robot and Human Interactive Communication, 2014 RO-MAN: The 23rd IEEE International Symposium on.

Electromyography. from http://en.wikipedia.org/wiki/Electromyography Electromyography (EMG). from http://www.diybiomechanics.com/wp-

content/uploads/2012/01/EMG.pdf Klug, S., Von Stryk, O., & Mohl, B. (2006). Design and control mechanisms for a 3 dof bionic

manipulator. Paper presented at the Biomedical Robotics and Biomechatronics, 2006. BioRob 2006. The First IEEE/RAS-EMBS International Conference on.

Layton, J. How can someone control a machine with her thoughts? , from http://science.howstuffworks.com/bionic-arm.htm

Made How. (2014). Mark W. Spong, S. H., and M. Vidyasagar. (2006). Robot Dynamics and Control (Robot Modeling and

Control) (Second ed.): John-Wiley and Sons New York. Medlej, J. (2014). Human Anatomy Fundamentals: Flexibility and Joint Limitations. Michael J. Quigley, C. P. O. (1988). Prosthetist's Role And Practice. Journal of the American Orthotic

and Prosthetic Association, 42. Moehl, B. (2000). Bionnic Robot arm with compliant actuators. Paper presented at the Sensor Fusion

and Decentralized Control in Robotic Systems III. Nature. (2014). from http://www.nature.com Ohol, S. (2014). KINEMATIC ANALYSIS TO GENERATE TRAJECTORY FOR SIX DEGREE OF FREEDOM

HUMANOID ROBOT ARM. IJITR, 2(1), 722-725. Oliver, R. G. E. C. K. E. B. a. C. W. (2011). Bionic prosthetic hands: A review of present technology and

future aspirations, Royal College of Surgeons of Edinburgh. 9(6), 336-340. Pavic, M. (2010). New Bionic Arms Are Strong, Sensitive, Human-Friendly. from

http://www.wired.com/2010/05/new-bionic-arms/ Robertson, G., Caldwell, G., Hamill, J., Kamen, G., & Whittlesey, S. (2013). Research Methods in

Biomechanics, 2E: Human Kinetics. Shekhar, H., Guha, R., Juliet, A. V., & Kumar, J. (2009). Mathematical Modeling of Neuro-Controlled

Bionic Arm. Paper presented at the Advances in Recent Technologies in Communication and Computing, 2009. ARTCom'09. International Conference on.

Surgeons, A. A. o. O. (2014). AAOS Online Newsroom. from http://newsroom.aaos.org Wallace, G. S. a. G. (2007). Conducting polymers – bridging the bionic interface. Soft Matter, 3, 65-

671. WORLDINFO. (2014). English Word Information. from http://wordinfo.info Zajac, F. E. (1988). Muscle and tendon: properties, models, scaling, and application to biomechanics

and motor control. Critical reviews in biomedical engineering, 17(4), 359-411.