Brain computer interfacing for controlling wheelchair movement

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 07 | July 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 2916
Brain computer interfacing for controlling wheelchair movement
Reshma Suresh1& Mrs. Dr. Subha Hency Jose, M.E., Ph.D2
1 2nd Year M.Tech Student, Karunya Institute of technology and Sciences
2 Associate Professor Karunya Institute of technology and Sciences
--------------------------------------------------------------------------***--------------------------------------------------------------------------
Abstract — The inability is extreme to the point that they
can't have any sort of developments. Confronted with the
present circumstance, Brain PC Interface innovation has
responded to the call of creating arrangements that permit
conveying a superior personal satisfaction to those
individuals, and quite possibly of the main region has been
the versatility arrangements, which incorporates the mind
PC interface empowered electric wheelchairs as perhaps of
the most supportive arrangement. Confronted with
everything going on, the current work has fostered a Brain
PC Interface arrangement that permits clients to control
the development of their wheelchairs utilizing the mind
waves created at the point when flickers their eyes. For the
production of this arrangement, the Steady Prototyping
approach has been utilized to improve the advancement
interaction by creating autonomous modules. The
arrangement is comprised of a few parts for example EEG
System (OpenBCI), Main Controller, Wheelchair Controller
and Wheelchair that permits to have a measured quality to
do refreshes (upgrades) of their functionalities in a basic
way. The created framework has shown that it requires a
low measure of preparing time and has a genuine material
reaction time. Exploratory outcomes demonstrate the way
that the clients can perform unique undertakings with an
OK grade of mistake in a timeframe that could be thought
of as OK for the framework. Considering that the model
was made for individuals with handicaps, the framework
could concede them a specific degree of freedom.
Key Words: alpha waves; brain–computer interface
(BCI); electroencephalography (EEG); wheelchair.
I INTRODUCTION
Brain–computer interface (BCI) is an immediate
correspondence way between the cerebrum and the outer
gadget [1-3], intended to investigate continuous mind
information to control a PC, neuroprosthesis, or
wheelchairs. Dissimilar to regular interfaces incorporating
the sign related with eye development
(electrooculography, EOG) [4, 5] or the facial muscle
withdrawals (electromyography, EMG) [6], BCI does not
need the association with muscles or fringe nerves, which
permits to control devices without verbal or actual
association [7-9]. This empowers patients in extreme
phases of disease that forestall any development, like
subcortical cerebral stroke, amyotrophic parallel sclerosis,
cerebral paralysis to impart with the rest of the world.
Generally, BCI frameworks depend on properties of
electromagnetic rushes of the cerebrum, recorded utilizing
electroencephalographic procedures [10-12]. In this
setting, the main issue is to record and investigate the
human produced electroencephalographic signs and then
make an interpretation of it into the machine control
succession. The probability of utilizing BCI to control a
wheelchair is wanted by patients, which brought about the
plan of numerous models of such BCI-based frameworks
[13]. The most straightforward one controls the
wheelchair that moves just in one bearing [14]. In that
study, the spinal rope harmed subject had the option to
produce explosions of beta motions in the EEG signal by
imagination of developments of his deadened feet. The
beta motions were utilized for an independent cerebrum
PC interface control based on a solitary bipolar recording.
The subject was put in a virtual road to mimic driving of
wheelchair prior to utilizing BCI in a genuine
circumstance. The BCI control system to drive a shrewd
wheelchair which allows the client to choose one of four
orders is proposed in Ref. [15]. When an order is chosen,
the control system executes the chose order what's more,
simultaneously, screens the close to home condition of the
client. While the client is fulfilled, the order is executed; in
any case, the control framework stops the wheelchair. A
large number of the presently evolved frameworks utilize
the half breed cerebrum PC interfaces [18-22]. For
instance, Wang et al. [18] consolidate engine symbolism,
P300 possibilities and eye squinting to carry out forward,
in reverse and stop control of a wheelchair, while Cao et al.
[21] consolidate engine symbolism (MI)- based bio-signals
and consistent state visual evoked possibilities (SSVEPs) to
control the speed and course of a wheelchair
simultaneously. BCI interfaces readily utilize the EEG
signal because of its great time goal and low working
expenses. Notwithstanding, according to the perspective of
a solitary client, these expenses are still high, and the
establishment of numerous anodes awkward and badly

designed. A more accommodating type of EEG signal
enlistment are EEG headbands that have as of late showed
up on special. They are not difficult to gather, however
have a predetermined number of estimating anodes and
the recorded signal is more regrettable quality contrasted
with proficient recorders. We intended to add to bringing
BCI frameworks outside the research center so it very well
may be more open to patients, by planning a calculation
which can work in view of the sign from modest number of
cathodes and with a low sign to commotion proportion.
Our paper portrays a mind waves impelled wheelchair
idea which depends on two mental conditions of the
subject condition of unwinding and condition of
concentration. Our answer depends on the examination of
the EEG signal for the event of the alpha waves. The state
when the alpha waves are available is treated as a parallel
state and the subject chooses the development bearing
utilizing the word reference of paired groupings. The
Human Brain is constructed with average of 100 billion
neurons. The interaction between these neurons can be
represented as thoughts and emotional states of human
mind. Every interaction between these active neurons
creates a minimal electrical discharge, which creates
different amplitude and frequencies like alpha wave, beta
wave, theta wave & delta wave. This is why EEG devices
are used to detect the waveforms of different human
activities. EEG based BCIs have potential applications for
assisting paralyzed patients.
II RELATED WORK
A number of previous studies have been conducted to
analyze brain wave signals for different applications using
BCIs . EEG based BCIs have been used for analysing several
human activities including attention level with
corresponding brain wave signals . Some studies shows
deflection in brain wave signal due to eye blink and several
applications and aspects based on eye blink . Assistive
control system and wheelchair have been designed for
paralyzed and quadriplegic patients. These designs are
based on head movement and retina and ocular movement
which uses accelerometer and image processing technique
respectively. Moving head repeatedly is not convenient for
quadriplegic patients as it causes fatigue. Besides
accelerometer can not detect eye blink. Moreover, in image
processing of ocular and retina movement a convenient
light source is needed to be present always near the eyes
of the patient. This causes fatigue of the patient’s eye. In
both of these techniques, while moving with wheelchair
the patient can not move his head and eye arbitrarily.
Using EEG signals according to attention level can solve
these problems. The use of Neuro Sky Mind wave EEG
Headset is convenient for working with brain signals,
attention level and meditation. The difficult stage in
planning BCI is in removing the highlights from the EEG
signals. The important highlights technique respect high
exactness in the order. For example, the work by Samraj et
al. [2], utilized versatile recursive band pass channel and
autoregressive displaying as another strategy for
separating the elements. The dataset was given by the
division of clinical informatics, University of Graz Signals
from C3 and C4 cathodes were utilized as the prevailing
signals from the EEG. The examination to gauge the
information was performed on a 24-years of age single
female unwinding on a seat. The extricated highlights was
utilized as a contribution for machine figuring out how to
group the elements to left and right engine symbolism.
Support Vector Machines (SVM) and Linear Discriminant
Analysis (LDA) are utilized as order techniques, the level of
blunder was gone between 17.1% to 24.2%. Li et al. [7]
performed probe 8 subjects, utilizing Muse headset.
Utilizing Common Space Pattern calculation (CSP) for
separating highlights, of left/right-hand development
symbolism, then, at that point, by applying SVM to order
the result. The trial was done two times effectively, first
with C3 and C4 sensor terminals with exactness of 90%,
and second, by adding gamma wave information from F7
and F8 cathode sensors, which work on the exactness from
90% to 95.1%. One more execution of BCI is accounted for
in [8], where a servo engine control was finished utilizing
EEG signals utilizing Emotiv Epoc signal and F3 anode
sensor channel. The antiquity commotion was eliminated
and the signs was separated to get the alpha recurrence
band (9-13 Hz) as an element extraction. The servo settled
to turn 90ι when it gets a heartbeat from the PC, and
afterward pivot 90ι back when gotten the second beat. The
examination applied on one individual and the level of
precision was not announced. The work detailed by Rani et
al. [9] depends on consideration level what's more, eye
flickering solidarity to control Robotic wheelchair. EEG
furthermore, EMG signals were caught utilizing Neyro Sky
Headset and signals are separated for the scope of Alpha
(8-12 Hz) and Beta (13-30 Hz) in Matlab. The strength saw
of the flicker level range from 0 to 255, zero methods little
eye flicker and 255 implies enormous eye flicker. Also, for
the consideration, 40 to 60 implies regular consideration,
from 61 to 80 methods marginally raised also, greater than
80 implies that raised. The qualities and the grades of
consideration and eye flickering were planned to explicit
bearing. The quantity of workers for the trial and
exactness were not referenced.

III OVERVIEW OF OUR SOLUTION
The filtered EEG data is converted to frequency domain to
estimate the Power Spectrum Density (PSD). The current
EEG data is compared with the EEG reference data. If the
reference data is smaller than the current data then, it
means the user is not focusing in any direction, hence, the
control signal is to stop the wheelchair. Otherwise, the step
which follows is to extract the features by Mu and Beta
frequency bands. Then, extracted features represent input
for the machine learning using SVM algorithm, which will
predict the output, to be transferred to the wheelchair
wirelessly and control it into the different directions.
Finally, taking new EEG signals and repeat the procedure.
Fig. 1 shows the system block diagram of components and
interconnection. It consists of an Arduino microcontroller
to control the wheelchair to the required direction. The
wheelchair is connected to EEG, Pulse sensor,lm35 The
motor driver shield is an output, as it receive directions via
the controller module to controls the motors. Fig. 2 shows
some hardware details of the wheelchair control circuit.
Fig-1 Block diagram of the EEG-based BCI system main
components and sensors.
Fig 2: Circuit Diagram of the control circuit of the
Wheelchair.
The designed system consists of two main parts, the
transmitter side, which sends the human brain thought
command, and the receiver side is the wheelchair with its
control unit.
III PROPOSED ALGORITHMS
A.EEG data recording rule
The subject envisioned a sum of 20 developments. The
trial directed in a tranquil spot and the method of the trial
is as follows: The subject sets quiet on a seat and attempt
to restrict his body developments previously and during
recording EEG. The subject likewise is approached to
restrict his eyes development to decrease the eye curios.
Then the subject thinks and envision outwardly moving a
pen as it was by all accounts a development with his arms.
The subject educated by a program on the PC, which shows
the guidance on the screen for synchronizing the occasions
of all investigation preliminaries.
B. EEG data recording method
For the training, the volunteer will think for four seconds
of each direction and in the same time should satisfied the
rules. Then, rest for 3 seconds, and repeat once more.
Finally, it will indicate the end of recording EEG data by a
message box. EEG data, shown in Figure 3, is saved for four
different directories in an Excel file formats from all 14
electrodes.
Fig 3: EEG data acquisition in time domain
IV.RESULTS
As it was mentioned earlier, EPOC headset does not
provide standard sensors for C3 and C4 electrodes,
therefore, the solution is to extract the frequency band
related to motor brain movement that will give a high
accuracy. Alpha and Beta bands provide a good

discrimination for the motor imagery. Hence, to select the
best classifier model we test two specific frequency bands:
Mu and Beta frequency bands (8 to 30 Hz) and Theta to
Beta frequency bands (4 - 30) Hz. For the first frequency
bands Mu to Beta, SVM classifier has the highest accuracy
with 79.2 % as shown in Table 1. For other frequency
bands the frequency ranged between 70% and 75%.
Table 1: Percentage of accuracy through different
frequency bands and classifiers
This result shows that the Mu and Beta frequency bands
are the most suitable frequency bands to extract the
features for the motor imagery movement. Whenever
decreasing the number of hidden layers, the performance
of the network design will increase, the highest
performance is 0.4412 with 3 hidden layers (Fig. 4).
Fig 4: Neural Network Pattern Recognition
V CONCLUSION
We propose a wheelchair constrained by EEG cerebrum PC
interface completely constrained by alpha mind waves,
utilizing the peculiarity of unwinding, which can work on a
sign with a low sign to-clamor proportion and a not many
anodes. It depends on the condition of unwinding and a set
of eight paired words that permit to push ahead, in
reverse, turn both ways, pivot 45° right and left as
well as to speed up movement. The application empowers
observing and controlling of electric wheelchair by the
career of the deadened individual. It is feasible to right the
subject's track or to close down the whole framework in
case of a perilous circumstance. Our tests performed on
three subjects uncovered high awareness of the proposed
BCI-framework with any preparation stage. Our answer is
thusly a basic and compelling technique to control a
wheelchair by individuals with engine paresis, which
doesn't require the use of many cathodes restricting the
patient's development or different reiterations of picture
assignments which a little while later make the patient
tired.
REFERENCES
[1] Kaitlyn Casimo; Kurt E. Weaver; Jeremiah Wander;
Jeffrey G. Ojemann, “BCI Use and Its Relation to Adaptation
in Cortical Networks”, IEEE Transactions on Neural Systems
and Rehabilitation Engineering, vol. 25, no. 10, pp. 1697–
1704, 2017.
[2] (2017) The Benjamin Kofi Prince Website.
[Online].Available:http://www.benjaminking.ca/quadriple
gic
[3] (2017) World Health Organization Website.
[Online].Availablehttp://www.who.int/mediacentre/facts
heets/fs358/en
[4] (2017) Disabled World Website. [Online].
Available:https://www.disabledworld.com/definitions/pa
raplegia
[5] M. Abu-Alqumsan, F. Ebert, A. Peer, “Goal-recognition-
based adaptive brain-computer interface for navigating
immersive robotic systems”,Journal of Neural Eng., vol. 14,
no. 3, 2017.
[6] M. Eid, A. Fernandez, “ReadGoGo!: Towards real-time
notification on readers' state of attention”, in Proc. of IEEE
XXIV International Conference on Information,
Communication and Automation Technologies (ICAT), 2013,
pp. 1-6.
[7] F. Karimi, J. Kofman, N. Mrachcz-Kersting, D. Farina, J.
Ning, “Comparison of EEG spatial filters for movement
related cortical potential detection”, in Proc. of IEEE 38th
Annual International Conference of the IEEE Engineering in
Medicine and Biology Society (EMBC), 2016, pp. 1576-1579.
[8] Ramesh C. R., Layla B. Das, “Brain Computer Interface
device for speech impediments” in Proc. of IEEE
International Conference on Control Communication &
Computing India (ICCC), 2015, pp. 349-352.
[9] N. H. Liu, C. Y. Chiang, H. C. Chu, “Recognizing the
degree of human attention using EEG signals from mobile
sensors”, Sensors , vol. 13, no. 8, pp. 10273–10285, 2013.

[10] B. Van Hal, S. Rhodes, B. Dunne, R. Bossemeyer, “Low-
cost EEGbased sleep detection” in Proc. of 36th Annual
International Conference of the IEEE Engineering in
[11] D. V. Poltavski, “The use of single-electrode wireless
EEG in biobehavioral investigations”, Methods in Molecular
Biology, vol. 1256, pp. 375-390, 2015.
[12] D. V. Poltavski, D. Biberdorf, T. V. Petros,
“Accommodative response and cortical activity during
sustained attention”, Vision Research, vol. 63, pp. 1-8, June
2012.
[13] M. Abo-Zahhad, S. M. Ahmed, S. N. Abbas, “A Novel
Biometric Approach for Human Identification and
Verification Using Eye Blinking Signal”, IEEE Signal
Processing Letters, vol. 22, no. 7, pp. 876-880,2015.
[14] Yubing Jiang; Hyeonseok Lee; Gang Li; Wan-Young
Chung, “High performance wearable two-channel hybrid
BCI system with eye closure assist” in Proc. of 38th Annual
[15] A. M. S. Ang; Z. G. Zhang; Y. S. Hung; J. N. F. Mak., “A
user-friendly wearable single-channel EOG-based human-
computer interface for cursor control in Proc. of IEEE 7th
International IEEE/EMBS Conference on Neural
Engineering (NER),2015, pp. 565-568.
[16] M. Varela, “Raw EEG signal processing for BCI control
based on voluntary eye blinks”, in Proc. of IEEE Thirty Fifth
Central American and Panama Convention (CONCAPAN
XXXV), 2015, pp. 1-6.
[17] D. Szibbo, A. Luo, T. J. Sullivan, (2012). “Removal of
blink artifacts in single channel EEG” in Proc. of Annual
[18] D. Puanhvuan, S. Khemmachotikun, P. Wechakarn, B.
Wijarn, Y. Wongsawat, “Navigation-synchronized
multimodal control wheelchair from brain to alternative
assistive technologies for persons with severe disabilities”,
Cognitive Neurodynamics, vol. 11, no. 2, pp. 117-134, April
2017.
[19] M. F. Ruzaij Al-Okby; S. Neubert; N. Stoll and K.
Thurow, “Development and testing of intelligent low-cost
wheelchair controller for quadriplegics and paralysis
patients”, in Proc. of IEEE 2nd International Conference on
Bio-engineering for Smart Technologies (BioSMART), 2017,
pp. 1-4.
[20] M. F. Ruzaij ; S. Neubert; N. Stoll and K. Thurow,
“Design and implementation of low-cost intelligent
wheelchair controller for quadriplegias and paralysis
patient”, in Proc. of IEEE 15th International Symposium on
Applied Machine Intelligence and Informatics (SAMI), 2017,
pp. 399-404.
[21] Imteyaz O. Qamar; Bashar A. Fadli; Ghazi Al Sukkar;
Musa Abdalla, “Head movement based control system for
quadriplegia patients”, in Proc. of IEEE 10th Jordanian
International Electrical and Electronics Engineering
Conference (JIEEEC), 2017, pp. 1-5.
[22] Bryce O’Bard; Alex Larson; Joshua Herrera; Dominic
Nega; Kiran George , “Electrooculography Based iOS
Controller for Individuals with Quadriplegia or
Neurodegenerative Disease”, in Proc. of IEEE International
Conference on Healthcare Informatics (ICHI), 2017, pp. 101-
106.
[23] M. F. Ruzaij ; S. Neubert; N. Stoll and K. Thurow,
“Multi-sensor robotic-wheelchair controller for handicap
and quadriplegia patients using embedded technologies”,
in Proc. of IEEE 9th International Conference on Human
System Interactions (HSI), 2016, pp. 103-109.
[24] A. J. Machado; J. F. Amador; M. J. Coello, “Wheelchair
control system for quadriplegics and ocular keyboard”, in
Proc. of IEEE Thirty Fifth Central American and Panama
Convention (CONCAPAN XXXV), 2015, pp.1-5.
[25] T. A. Izzuddin; M. A. Ariffin; Z. H. Bohari; R. Ghazali; M.
H. Jali, “Movement intention detection using neural
network for quadriplegic assistive machine”, in Proc. of
IEEE International Conference on Control System,
Computing and Engineering, 2015, pp. 275-280.
[26] Umar Mohammad; Mohammad Anas, “Design of a low
cost DIY moving wheel chair using ATmega1284P based
on retina movement for the persons disabled with
quadriplegia”, in Proc. of Annual IEEE India Conference
(INDICON), 2015, pp. 1-4.
[27] Jozsef Katona; Tibor Ujbanyi; Gergely Sziladi; Attila
Kovari, “Speed control of Festo Robotino mobile robot
using NeuroSky MindWave EEG headset based brain-
computer interface”, in Proc. of 7th IEEE International
Conference on Cognitive Infocommunications (CogInfoCom),
2016, pp. 251-256.

[28] MindWave Mobile : User Guide, NeuroSky Inc., August
2015.
[29] NeuroExperimenter Users’ Guide, NeuroSky Inc.,
September 2015.

Brain computer interfacing for controlling wheelchair movement

More Related Content

Similar to Brain computer interfacing for controlling wheelchair movement (20)

More from IRJET Journal (20)

Recently uploaded (20)

Brain computer interfacing for controlling wheelchair movement