ABSTRACT

The concept of brain computer interface (BCI) was introduced in late 80’s and has been
actively developed over the last decades. The core objective of the BCI system is to
develop a control system that will be able to communicate with the brain thoughts.
Electroencephalography (EEG) captured from the human scalp is commonly used an
information carrier for most of the BCI system. Due to less power of the EEG signals
captured from the human scalp which makes it a difficult task to achieve but as compared
to other technologies like Electromyography (EMG) it is able to produce signals with
more accuracy. The objective of this project is to develop a prosthetic hand to assist
disable person using EEG based BCI algorithm which will be able to identify the physical
movements of each and every finger joint and by using this EEG technology an algorithm
will be created which is more than 75% accurate i.e. whatever physical movement a
disable person think will be done using it.

1
ACKNOWLEDGEMENT

We take this opportunity to express our sincere gratitude to our thesis advisor, Prof. Tahir
Hussain Rizvi, Associate Professor, Department of Electrical Engineering, University of
Central Punjab and co-advisor, Prof. Yaqoob Javed, Associate Professor, Department of
Electrical Engineering, University of Central Punjab. We thank them for guiding us in
selecting this area of research from which we are also able to work on it in the future. I
could not have imagined having a better advisor and co-advisor for my study and
research.

We are also thankful to our department chair, Prof. Azhar Hussain Shah, for his guidance
and support.

We would like to thank our parents as they are constant source of inspiration. We would
like to delicate this project work to our parents because After The God they are the
greatest support in our life’s.

2
 CHAPTER: 1

INTRODUCTION

We are living in a planet of change and this change will not be possible without latest
technology. As the technology increases scientist are trying to make human movements
in prosthetic devices using neuro-signals. These changes often provide tools for
improving living conditions and facilitating mankind. Advantages in the field of
neuroscience are also remarkable and Electroencephalography (EEG) is one of them.
This technology allows viewing and analyzing the electrical impulses of the brain
through electrodes placed on a person’s head. EEG provides a view of signals produced
by brain to perform various activities. Captured signal can be used to perform various
tasks like control of Prosthetic devices (Leg, hand, and fingers etc), gaming and control
of different facilities (wheel chair, devices etc) through EEG signals.

The most important advantage of EEG over other techniques like Electromyography
(EMG) is that they require muscular stimuli, which may not be generated in patients
suffering severe paralysis like amyotrophic lateral sclerosis (ALS), brainstem stroke etc.
So, Patients suffering severe muscular defects can utilize their brain stimuli (EEG Signal)
to control any system or prosthetic device [1].Due to this major advantage of EEG many
engineers and bio-medical science students are working to use this technology more and
more effectively.

In our project, we focus on controlling the prosthetic hand with brain thoughts usually
known to be EEG (Electroencephalography) technology in medical science. Two types of
controlling systems can be made invasive and non-invasive systems. In invasive systems
the sensors used for the detection of signals are inserted into the brain which makes it
more accurate with high SNR value. It may cause physical harm to the patient and
required more accuracy to insert so, here we are using the non-invasive which can be

3
easily placed on the brain scalp. Here we use fourteen electrodes and two electrodes as
and reference so that the headset can be easily placed on the brain scalp which is able of
recording the brain signals on real time. These signals will be amplified and then
transferred to SBC (Single Board Computers) where the captured signals will be filtered
and analyzed to classify the EEG signals and detect movement of fingers, hands to be
controlled. Any prosthetic device can be controlled using this project. Main idea to use
this methodology is to reduce the system complexity. Our system will be portable, light
weight and user friendly small training will be needed to use the device

1.1 Literature Survey:

Since the early 1900s and has had much interest ever since [3]. Although the first
documented human brain activity was recorded in 1929 and it was named as EEG, after
this discovery, EEG was primarily used for assistance in clinical diagnostics. In 1973
Jacques J. Vidal from University of California introduces the concept of “Brain
Computer Communication” he proposed the concept of Visual Evoked Potential and the
route for the future of Brain Computer Interface [1]. For today’s wounded warriors
DARPA’s Revolutionizes Prosthetics program set out to expand prosthetic arm options. 
The program funded two teams to create advanced anthropomorphic mechanical arms
and control systems; one to get an arm control system to market quickly, the other to
determine the viability of direct brain control [2]. EEG has been most predominantly
researched for medical applications, with a large history of use relating to epilepsy. The
DAPRA institute has just build the prosthetic hand while using EEG signals which is a
million dollar project [2]. It was difficult to discriminate all the finger motions from just
EMG [4]. With advances in technology and research, it has become possible to process
EEG data in real-time. By detecting features of the brain activity and creating a feedback
loop, users can communicate to a computer without the need for any physical input [5].
Major advancement in gaming is to control the object using brain signals and brain

4
controlled wheel chairs have also been made. This advancement tends us to work on EEG
technology.

After collecting data from different research work we concluded our project to be user
friendly and to make it less complex. Our project of prosthetic hand controlling system
will not have that much freedom of movements like a human hand but try to make a
prosthetic hand which can help disabled people in some proportion.

1.2 OBJECTIVES

Robotic arms can be implemented for multiple applications in industry, from welding
material handling, thermal spraying to painting and drilling. DAPRA robotic arm is one
of them. The robotic hand which we are trying to build in our final year project performs
the function of grasping with movement of fingers. This robotic hand can be used to
facilitate disabled especially the brain stroke patients whose muscles have also been
disabled. Main objective of our final year project is to make a BCI system which can be
able to communicate with any prosthetic device.

5
1.3 BLOCK DIAGRAM

WIRELESS
EMOTIV –
SIGNALS
EPOC SCALP BRAIN WIRELESS Tx
HEADSET SIGNAL

SINGLE BOARD COMPUTER [SBC]

DIGITAL SIGNAL PROCESSING

MOTOR DRIVING ROBOTIC

Fig. 1-3: Block diagram

6
1.3.1 Methodology:

Emotive EPOC headset will be used which consists of fourteen electrodes that will be
used to capture EEG signals from the brain scalp and two electrodes as an reference so
that the headset (electrodes) can be easily placed on the specific places these raw brain
signals will be then wirelessly transmitted with the help of dongle USB which will be
attached to SBC where the recorded signals from the brain scalp will be processed to
filter out the required signal with notch and 5th order sink filter to increase the signal to
noise ratio so that the task to be performed should be with greater efficiency. These
signals will be then serially transmitter to the microcontroller which will be able to
control the movements of the prosthetic hand or any prosthetic device which will be
attached to it.